# sql_to_fastapi_LLM_generator

**Repository Path**: ncepu-bj/sql_to_fastapi_LLM_generator

## Basic Information

- **Project Name**: sql_to_fastapi_LLM_generator
- **Description**: 通过大模型来生成基于fastapi框架的标准RESTful风格的分层接口项目
- **Primary Language**: Python
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 3
- **Forks**: 0
- **Created**: 2025-08-27
- **Last Updated**: 2026-01-10

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# SQL to FastAPI LLM Code Generator

<div align="center">

![Python](https://img.shields.io/badge/python-v3.9+-blue.svg)
![FastAPI](https://img.shields.io/badge/FastAPI-0.100+-green.svg)
![LLM](https://img.shields.io/badge/LLM-DeepSeek%20|%20Qianwen%20|%20Kimi-red.svg)
![License](https://img.shields.io/badge/license-MIT-blue.svg)

**Intelligent SQL to FastAPI Code Generator powered by Chinese Large Language Models**

*Transform SQL table structures into high-quality FastAPI RESTful API projects through enhanced prompt engineering*

</div>

## 🎯 Overview

This project is an intelligent code generator powered by Large Language Models (LLMs) that transforms SQL database table structures into complete FastAPI RESTful API projects. The project adopts a core strategy of "**Enhanced Prompt Engineering + Lightweight Templates**", supports mainstream Chinese LLMs (DeepSeek, Qianwen, Kimi), and generates complete three-tier architecture code including model layer, service layer, and API layer.

### ✨ Key Features

- 🧠 **Intelligent Prompt System**: Deep analysis of SQL table structures to generate targeted professional prompts
- 📊 **Smart Complexity Assessment**: Multi-dimensional evaluation of table structure complexity with automatic optimal strategy selection
- 🎯 **Layered Code Generation**: Complete generation of SQLAlchemy models, business services, and FastAPI routes
- 🔍 **Code Quality Assurance**: Built-in syntax checking, quality scoring, and improvement suggestion system
- 🌏 **Chinese LLM Optimization**: Specially optimized for DeepSeek, Qianwen (Tongyi Qianwen), Kimi and other domestic models
- 🏗️ **Enterprise Architecture**: Standard three-tier architecture with complete exception handling and data validation

## 🏗️ System Architecture

```
SQL Structure → Context Analysis → Complexity Assessment → Smart Prompts → LLM Generation → Code Validation → FastAPI Project
      ↓              ↓                    ↓                ↓              ↓               ↓                ↓
   Parser      Semantic Inference    Strategy Selection   Layer Prompts   Model Call   Quality Check   Project Assembly
```

### Core Components
- **SQL Parser**: Parse SQL DDL and extract table structures and constraint information
- **Prompt System**: Intelligently analyze and generate high-quality code generation prompts
- **LLM Integration**: Unified interface supporting multiple Chinese large language models
- **Code Generation Engine**: Generate complete FastAPI project code based on analysis results
- **Quality Validator**: Ensure syntax correctness and best practices of generated code

## 🚀 Quick Start

### Requirements

- Python 3.9+
- Supported LLM Services: DeepSeek API, Qianwen API, Kimi API

### Installation

1. **Clone the repository**
```bash
git clone https://github.com/your-repo/sql_to_fastapi_LLM_generator.git
cd sql_to_fastapi_LLM_generator
```

2. **Create virtual environment**
```bash
# Using conda (recommended)
conda create -n sql_to_fastapi python=3.11
conda activate sql_to_fastapi

# Or using venv
python -m venv venv
source venv/bin/activate  # Linux/Mac
# or venv\Scripts\activate  # Windows
```

3. **Install dependencies**
```bash
pip install -r requirements.txt
```

4. **Configure LLM services**
```bash
# Copy configuration template
cp .env.example .env

# Edit .env file and add your LLM API keys
# DEEPSEEK_API_KEY=your_deepseek_api_key
# QIANWEN_API_KEY=your_qianwen_api_key  
# KIMI_API_KEY=your_kimi_api_key
```

### Basic Usage

1. **Prepare SQL file**
```sql
-- Save your SQL DDL to input/ directory
-- Example: input/my_database.sql
CREATE TABLE users (
    id INT PRIMARY KEY AUTO_INCREMENT,
    username VARCHAR(50) NOT NULL UNIQUE,
    email VARCHAR(100) NOT NULL,
    password VARCHAR(255) NOT NULL,
    created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
    updated_at DATETIME ON UPDATE CURRENT_TIMESTAMP
);
```

2. **Run generator**
```bash
python start.py --input input/my_database.sql --output output/my_project
```

3. **Check generated results**
```
output/my_project/
├── app/
│   ├── models/          # SQLAlchemy models
│   ├── services/        # Business logic layer
│   ├── routers/         # FastAPI routes
│   └── schemas/         # Pydantic data models
├── requirements.txt     # Project dependencies
└── main.py             # FastAPI application entry
```

## 📚 Documentation

### Core Documentation
- 📋 [Implementation Roadmap](./docs/IMPLEMENTATION_ROADMAP.md) - Overall project planning and progress
- 🏗️ [Prompt System Architecture](./docs/prompt_system_architecture.md) - Core prompt system detailed explanation
- 📖 [Prompt Engineering Guide](./docs/prompt_engineering_guide.md) - Prompt design and optimization

### Development Documentation  
- 🎯 [Week 2 Summary](./docs/WEEK2_SUMMARY.md) - Latest development progress
- 🏢 [Enterprise Architecture Design](./docs/enterprise_architecture_design.md) - Overall architecture design
- 📝 [Documentation Index](./docs/README.md) - Complete documentation directory

## 🧪 Testing

Run complete test suite:
```bash
# Run all tests
python -m pytest tests/ -v

# Run specific tests
python -m pytest tests/test_prompt_system.py -v
python -m pytest tests/test_llm_clients.py -v
```

## 🎨 Project Highlights

### 🧠 Intelligent Prompt Engineering
- **Business Semantic Analysis**: Automatically infer field business meanings (email, phone, password, etc.)
- **Complexity Adaptation**: Intelligently select generation strategies based on table complexity
- **Layer Specialization**: Generate specialized prompts for model, service, and API layers

### 🎯 Code Quality Assurance
- **Syntax Validation**: Automatically check syntax correctness of generated code
- **Quality Scoring**: Multi-dimensional quality scoring of generated code (1-10 points)
- **Best Practices**: Built-in FastAPI and SQLAlchemy best practices

### 🌏 Chinese LLM Deep Optimization
- **DeepSeek**: Prompt strategies optimized for DeepSeek models
- **Qianwen (Tongyi Qianwen)**: Special handling adapted for Alibaba Cloud Qianwen models
- **Kimi**: Support for Moonshot Kimi model's long-text capabilities

## 📊 Development Progress

### ✅ Phase 1 (Completed)
- [x] Project basic architecture
- [x] SQL parser
- [x] Database schema analyzer
- [x] LLM client integration
- [x] Exception handling system

### ✅ Week 2 (Completed) - Prompt System Framework
- [x] Intelligent prompt builder (227 lines of code)
- [x] Deep context analyzer (398 lines of code)
- [x] Multi-dimensional complexity evaluator (414 lines of code)
- [x] Lightweight template system (699 lines of code)
- [x] LLM response parser (502 lines of code)
- [x] Complete test suite (2574 lines total code)

### 🔄 Week 3 (In Progress) - Generation Engine Core
- [ ] Generation engine master controller
- [ ] Simple CRUD generation strategy
- [ ] Basic code validator
- [ ] Project assembler
- [ ] Basic CLI commands

## 🤝 Contributing

We welcome all forms of contributions!

### Ways to Contribute
1. 🐛 **Report Bugs**: Describe issues in Issues
2. 💡 **Feature Suggestions**: Propose new feature ideas
3. 📖 **Improve Documentation**: Enhance project documentation
4. 💻 **Code Contributions**: Submit Pull Requests

### Development Workflow
```bash
# 1. Fork this repository
# 2. Create feature branch
git checkout -b feature/your-feature-name

# 3. Commit changes
git commit -m "Add: your feature description"

# 4. Push to branch
git push origin feature/your-feature-name

# 5. Create Pull Request
```

## 📜 License

This project is licensed under the [MIT License](LICENSE).

## 🙏 Acknowledgments

- Thanks to [DeepSeek](https://deepseek.com/) for providing excellent large language model services
- Thanks to [Alibaba Cloud Tongyi Qianwen](https://tongyi.aliyun.com/) for technical support
- Thanks to [Moonshot Kimi](https://kimi.moonshot.cn/) for long-text processing capabilities
- Thanks to the FastAPI and SQLAlchemy communities for their outstanding contributions

---

<div align="center">

**🌟 If this project helps you, please give us a Star!**

Made with ❤️ by SQL to FastAPI LLM Generator Team

</div>

1.  You can use Readme\_XXX.md to support different languages, such as Readme\_en.md, Readme\_zh.md
2.  Gitee blog [blog.gitee.com](https://blog.gitee.com)
3.  Explore open source project [https://gitee.com/explore](https://gitee.com/explore)
4.  The most valuable open source project [GVP](https://gitee.com/gvp)
5.  The manual of Gitee [https://gitee.com/help](https://gitee.com/help)
6.  The most popular members  [https://gitee.com/gitee-stars/](https://gitee.com/gitee-stars/)