LSPRAG - Language-Agnostic Real-Time Unit Test Generation

VS Code Extension for AI-Powered Unit Test Generation

Overview

LSPRAG (Language Server Protocol-based AI Generation) is a cutting-edge VS Code extension that leverages Language Server Protocol (LSP) integration and Large Language Models (LLMs) to automatically generate high-quality unit tests in real-time. By combining semantic code analysis with AI-powered generation, LSPRAG delivers contextually accurate and comprehensive test suites across multiple programming languages.

✨ Key Features

🚀 Real-Time Generation

• Generate unit tests instantly as you code
• Context-aware test creation based on function semantics
• Intelligent test case generation with edge case coverage

🌍 Multi-Language Support

• Java: Full support with JUnit framework
• Python: Comprehensive pytest integration
• Go: Native Go testing framework support
• Extensible: Easy to add support for additional languages

🎯 Advanced Capabilities

• Semantic Analysis: Deep code understanding through LSP
• Dependency Resolution: Automatic import and mock generation
• Coverage Optimization: Generate tests for maximum code coverage
• Multiple LLM Providers: Support for OpenAI, DeepSeek, and Ollama
• Customizable Prompts: Multiple generation strategies available

🎯 Project Status

Language	Status	Framework	Features
Java	✅ Production Ready	JUnit 4/5	Full semantic analysis, mock generation
Python	✅ Production Ready	pytest	Type hints, async support, fixtures
Go	✅ Production Ready	Go testing	Package management, benchmarks

🛠️ Installation & Setup

Prerequisites

• VS Code: Version 1.95.0 or higher
• Node.js: Version 20 or higher

Installation Steps

Note: Currently, LSPRAG is available only as source code. While we plan to publish it as a one-click extension in the future, we're maintaining source-only distribution to preserve anonymity. Please follow the steps below to set up the application.

1. Download Source Code

   • Use git clone or download the ZIP file directly

2. Setup Project

   • Navigate to the project's root directory LSPRAG
   • Install dependencies: npm install --force
   • Compile the project: npm run compile

3. Install Language Server Extensions

   For Python:

   • Install "Pylance" and "Python" extensions

   For Java:

   • Install "Oracle Java Extension Pack" from VS Code Marketplace

   For Go:

   • Install "Go" extension
   • Enable semantic tokens in settings:

   {
     "gopls": {
       "ui.semanticTokens": true
     }
   }

4. Download Baseline Python Project

   • Navigate to experiments directory: cd experiments
   • Create projects folder: mkdir projects
   • Clone a sample project: git clone https://github.com/psf/black.git

5. Activate Extension

   • Navigate to src/extension.ts
   • Click "Run and Debug" and select "VS Code Extension Development"
   • A new VS Code editor will open - use this for subsequent actions

6. ⚠️ IMPORTANT: Configure LLM Settings in the NEW Editor

   Critical: You must configure your LLM settings in the newly opened VS Code editor (not the original one) for LSPRAG to work properly.

   Option A: VS Code Settings UI

   • Open VS Code Settings (Ctrl/Cmd + ,)
   • Search for "LSPRAG" settings
   • Configure provider, model, and API keys

   Option B: Direct JSON Configuration Add to your settings.json:

   {
     "LSPRAG": {
       "provider": "deepseek",
       "model": "deepseek-chat",
       "deepseekApiKey": "your-api-key",
       "openaiApiKey": "your-openai-key",
       "localLLMUrl": "http://localhost:11434",
       "savePath": "lsprag-tests",
       "promptType": "detailed",
       "generationType": "original",
       "maxRound": 3
     }
   }

   Test your configuration with Ctrl+Shift+P → LSPRAG: Show Current Settings

🚀 Usage

Basic Workflow

1. Open Your Project

   • Open your workspace in the new VS Code editor
   • Navigate to the black project: LSPRAG/experiments/projects/black
   • Ensure language servers are active for your target language

2. Generate Tests

   • Navigate to any function or method
   • Right-click within the function definition
   • Select "LSPRAG: Generate Unit Test" from the context menu
   • Wait for generation to complete

3. Review & Deploy

   • Generated tests will appear with accept/reject options

Command Palette Commands

• LSPRAG: Generate Unit Test - Generate tests for selected function
• LSPRAG: Show Current Settings - Display current configuration
• LSPRAG: Test LLM - Test LLM connectivity and configuration

⚙️ Configuration

Core Settings

Setting	Type	Default	Description
LSPRAG.provider	string	"deepseek"	LLM provider (deepseek, openai, ollama)
LSPRAG.model	string	"deepseek-chat"	Model name for generation
LSPRAG.savePath	string	"lsprag-tests"	Output directory for generated tests
LSPRAG.promptType	string	"basic"	Prompt strategy for generation
LSPRAG.generationType	string	"original"	Generation approach
LSPRAG.maxRound	number	3	Maximum refinement rounds

API Configuration

DeepSeek

{
  "LSPRAG.provider": "deepseek",
  "LSPRAG.model": "deepseek-chat",
  "LSPRAG.deepseekApiKey": "your-api-key"
}

OpenAI

{
  "LSPRAG.provider": "openai",
  "LSPRAG.model": "gpt-4o-mini",
  "LSPRAG.openaiApiKey": "your-api-key"
}

Ollama (Local)

{
  "LSPRAG.provider": "ollama",
  "LSPRAG.model": "llama3-70b",
  "LSPRAG.localLLMUrl": "http://localhost:11434"
}

Advanced Features

Generation Strategies

• naive: Basic test generation without semantic analysis
• original: Standard LSP-aware generation (recommended)
• agent: Multi-step reasoning with iterative refinement
• cfg: Control flow graph-based generation
• experimental: Latest experimental features
• fastest: Optimized for speed
• best: Highest quality generation

Prompt Types

• basic: Minimal context, fast generation
• detailed: Comprehensive context analysis
• concise: Balanced approach
• fastest: Speed-optimized prompts
• best: Quality-optimized prompts

💻 System Requirements

Hardware Requirements

• Minimum: 8GB RAM, 4 CPU cores
• Recommended: 16GB RAM, 8 CPU cores
• GPU: Optional but recommended for local LLM inference

---

Ready to generate unit tests with LSPRAG! 🎉