Metadata-Version: 2.1
Name: querynl-cli
Version: 0.2.6
Summary: AI-powered CLI for natural language database queries and automated test data generation
Home-page: https://github.com/dushshantha/QueryNL
Author: QueryNL Team
Author-email: QueryNL Team <contact@querynl.io>
License: MIT
Project-URL: Homepage, https://github.com/dushshantha/QueryNL
Project-URL: Documentation, https://github.com/dushshantha/QueryNL/blob/main/README.md
Project-URL: Repository, https://github.com/dushshantha/QueryNL
Project-URL: Issues, https://github.com/dushshantha/QueryNL/issues
Keywords: database,sql,natural-language,cli,test-data,faker,llm,ai
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: System Administrators
Classifier: Topic :: Database
Classifier: Topic :: Software Development :: Code Generators
Classifier: Topic :: Utilities
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE

# QueryNL - Natural Language Database Queries from the Terminal

Query your databases using plain English. No SQL required.

[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
[![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)

## Features

- **Natural Language Queries**: Ask questions in plain English, get SQL results
- **AI-Powered**: Uses OpenAI GPT-4 or Anthropic Claude for intelligent query generation
- **Multi-Database**: PostgreSQL, MySQL, SQLite, MongoDB support
- **Schema Design**: Design database schemas through conversation
- **Test Data Generation**: Automatically populate tables with realistic data
- **Interactive REPL**: Tab completion, history, and context-aware queries
- **Secure**: Credentials stored in OS keychain, SSL/TLS by default
- **Safe**: Preview SQL before execution, confirmations for destructive operations

## Installation

### From PyPI (Coming Soon)

```bash
pip install querynl-cli
```

### From Source

```bash
pip install git+https://github.com/yourusername/QueryNL.git
```

### Prerequisites

- Python 3.10 or higher
- OpenAI or Anthropic API key (for AI features)

## Quick Start

### 1. Configure LLM Provider

```bash
# Using OpenAI
querynl config llm --provider openai --test

# Or using Anthropic Claude
querynl config llm --provider anthropic --test
```

### 2. Add a Database Connection

```bash
querynl connect add mydb
# Follow the interactive prompts
```

### 3. Start Querying

```bash
querynl repl

querynl (mydb)> show me all users created this month
querynl (mydb)> count orders by status
querynl (mydb)> find top 10 products by revenue
```

## Usage Examples

### Natural Language Queries

```bash
# In REPL mode
querynl repl

querynl> show me all active users
querynl> count orders grouped by status
querynl> find products with price > 100
querynl> update user email where id = 5
```

### Schema Design

Create database schemas through natural language conversation:

```bash
querynl> \schema design
Schema Designer> I need a blog with users, posts, and comments
Schema Designer> posts should have titles and content
Schema Designer> users can comment on posts
Schema Designer> finalize

# Execute the generated schema
querynl> \schema execute
```

### Test Data Generation

Populate your tables with realistic test data:

```bash
querynl> add sample data
querynl> add 100 users and 500 posts
querynl> add e-commerce product data
```

### Command-Line Mode

```bash
# Execute a single query
querynl query exec "count all users"

# Output as JSON
querynl query exec --format json "all active users"

# Save to file
querynl query exec --output data.csv --format csv "all orders"

# Preview SQL without executing
querynl query exec --explain "delete old records"
```

## Key Features

### 🤖 AI-Powered Query Generation

- Understands context and database schema
- Generates optimized SQL queries
- Supports complex joins, aggregations, and subqueries
- Confidence scoring for query accuracy

### 🔒 Security First

- Credentials stored in OS keychain (macOS, Windows, Linux)
- SSL/TLS enabled by default
- Preview SQL before execution
- Confirmation prompts for destructive operations (DELETE, DROP, etc.)
- No credentials in config files or logs

### 🎨 Beautiful Terminal UI

- Syntax-highlighted SQL
- Formatted table output
- Progress bars for long operations
- Tab completion for commands and tables
- Persistent command history

### 🌐 Multi-Database Support

- **PostgreSQL** - Full support
- **MySQL** - Full support
- **SQLite** - Full support
- **MongoDB** - Basic support

### 📊 Schema Design & Test Data

- Design schemas through natural language
- Generate DDL for any database
- View ER diagrams in terminal
- Populate tables with realistic test data
- Respects foreign keys and constraints

## Configuration

### Config File Locations

- **macOS**: `~/Library/Application Support/querynl/config.yaml`
- **Linux**: `~/.config/querynl/config.yaml`
- **Windows**: `%APPDATA%\querynl\config.yaml`

### Environment Variables

| Variable | Description |
|----------|-------------|
| `QUERYNL_CONNECTION_STRING` | Database connection string |
| `QUERYNL_KEYRING_PASSWORD` | Master password for headless servers |
| `QUERYNL_CONFIG` | Custom config file path |
| `QUERYNL_NO_COLOR` | Disable color output |

## REPL Commands

| Command | Description |
|---------|-------------|
| `\help` | Show available commands |
| `\tables` | List all tables in database |
| `\schema` | Display database schema |
| `\schema design` | Start schema design mode |
| `\schema execute` | Execute generated DDL |
| `\history` | Show query history |
| `\clear` | Clear screen |
| `exit` or `\quit` | Exit REPL |

## Examples

### E-commerce Analytics

```bash
querynl> show top 10 customers by total order value
querynl> count orders by status for last month
querynl> find products with low inventory
```

### User Management

```bash
querynl> show users who signed up this week
querynl> count active users by region
querynl> find users with no recent activity
```

### Data Cleanup

```bash
querynl> delete records older than 90 days from logs
querynl> update email domain from old.com to new.com
querynl> remove duplicate entries from customers
```

## Troubleshooting

### Connection Issues

```bash
# Test your connection
querynl connect test mydb

# List all connections
querynl connect list

# Remove and re-add connection
querynl connect remove mydb
querynl connect add mydb
```

### Keychain Access (macOS)

If you get "Permission denied to access keychain":

1. Open **Keychain Access** app
2. Search for "querynl"
3. Right-click → Get Info → Access Control
4. Add Terminal/iTerm to allowed applications

### Headless Servers (Linux)

```bash
# Use encrypted file keyring
export QUERYNL_KEYRING_PASSWORD="your-master-password"
querynl connect add mydb
```

## Documentation

- [LLM Setup Guide](LLM_SETUP.md)
- [Schema Design Quickstart](SCHEMA_DESIGN_QUICKSTART.md)
- [Developer Guide](DEV.md)
- [API Documentation](docs/)

## Support

- 🐛 [Report Issues](https://github.com/yourusername/QueryNL/issues)
- 💬 [Discussions](https://github.com/yourusername/QueryNL/discussions)
- 📧 Email: support@querynl.com

## License

MIT License - see [LICENSE](LICENSE) file for details

## Acknowledgments

Built with:
- [Click](https://click.palletsprojects.com/) - CLI framework
- [Rich](https://rich.readthedocs.io/) - Terminal formatting
- [LangChain](https://langchain.com/) - LLM orchestration
- [SQLAlchemy](https://www.sqlalchemy.org/) - Database toolkit
- [Faker](https://faker.readthedocs.io/) - Test data generation

---

**Made with ❤️ for developers who prefer English over SQL**
