You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

309 lines
11 KiB

# BRAIN Expression Template Decoder - Flask Web Application
A comprehensive Flask web application for decoding string templates with grammar checking, syntax highlighting, and seamless integration with WorldQuant BRAIN platform for real operator data.
## 🚀 Features
### Core Functionality
- **Template Detection**: Automatically detects `<template_name/>` patterns in expressions
- **Grammar Checking**: Real-time validation for semicolons, comments, and syntax
- **Syntax Highlighting**: Color-coded display with line numbers and autocomplete
- **Template Decoding**: Two modes for generating expressions
- **Full Iteration**: Generate all possible combinations of template variables
- **Random Iteration**: Generate a random sample of specified size
- **Export Options**: Copy results or download as text file
### Template Types
- **Op (Operators)** - Blue button, integrates with BRAIN operators
- **Data (Data Fields)** - Green button, for data field configuration
- **Normal (Parameters)** - Gray button, for general parameters
### BRAIN Integration
- **Built-in API Integration**: Direct connection to WorldQuant BRAIN API with CORS handling
- **186+ Operators**: Access to all operator categories (Arithmetic, Ranking, Reduce, Time Series, etc.)
- **800+ Data Fields**: From fundamental datasets with search and filtering
- **Advanced Search**: Filter operators by category and search by name
- **Multi-select Interface**: Professional operator selection with visual feedback
## 🛠 Setup and Installation
### Prerequisites
- Python 3.7 or higher
- pip (Python package installer)
- WorldQuant BRAIN account (for BRAIN integration features)
### Quick Start
1. **Clone or download** this repository to your local machine
2. **Run the application** (dependencies will be installed automatically!):
**On Windows:**
```bash
# Double-click run_app.bat or run in terminal:
run_app.bat
```
**On macOS/Linux:**
```bash
# Make the script executable (first time only):
chmod +x run_app.sh
# Run the application:
./run_app.sh
```
**Or run directly with Python:**
```bash
python app.py
```
3. **Automatic dependency installation**:
- The application will automatically check for missing dependencies
- If any packages are missing, they will be installed from requirements.txt
- **For users in China**: Dependencies are automatically downloaded from Tsinghua University mirror for faster installation
- No manual installation needed!
4. **Open the application**:
- The Flask server starts on `http://localhost:5000`
- Open your web browser and navigate to `http://localhost:5000`
- Everything is integrated - no separate proxy server needed!
### For Users in China (中国用户)
The application automatically uses Tsinghua University's PyPI mirror (清华大学镜像) for faster downloads. If you prefer to install dependencies manually:
**Windows:**
```bash
# Use the setup script with Tsinghua mirror
setup_tsinghua.bat
```
**macOS/Linux:**
```bash
# Make executable and run
chmod +x setup_tsinghua.sh
./setup_tsinghua.sh
```
**Manual installation with Tsinghua mirror:**
```bash
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt
```
**Configuring Alternative Mirrors:**
You can change the mirror source by editing `mirror_config.txt`. The file includes several popular Chinese mirrors:
- Tsinghua (清华大学) - Default
- Aliyun (阿里云)
- Douban (豆瓣)
- USTC (中国科学技术大学)
### Manual Installation
If the automatic installation doesn't work, you can install manually:
```bash
# Install dependencies from requirements.txt
pip install -r requirements.txt
# Or install individual packages
pip install flask==2.3.3 flask-cors==4.0.0 requests==2.31.0 pandas==2.0.3
# Start the application
python app.py
```
**Note**: The application includes automatic dependency checking and will attempt to install missing packages when you run it.
## 🔗 BRAIN Integration
### Setup
1. **No separate setup required** - BRAIN API integration is built into the Flask application
2. **Connect to BRAIN**: Click "Connect to BRAIN" button in the web interface
3. **Enter credentials**: Use your WorldQuant BRAIN username and password
4. **Automatic operator loading**: The system fetches all available operators (186+)
### Using BRAIN Operators
1. **Configure templates**: Set template type to "Op"
2. **Choose from BRAIN**: Click the "Choose from BRAIN" button in template configuration
3. **Search and filter**: Use the search box and category filter to find operators
4. **Multi-select**: Check multiple operators and apply them to your template
5. **Generate combinations**: Decode templates to get all possible operator combinations
## 📝 Usage Guide
### 1. Expression Editor
- Type or paste your expression in the main editor
- Grammar is checked automatically with visual feedback
- Templates are detected and highlighted in real-time
- Use autocomplete by typing `<` for template suggestions
### 2. Template Management
- Templates appear in the right panel as they're detected
- Click template type buttons (Op/Data/Normal) to configure
- Visual indicators show configured (✓) vs unconfigured (•) templates
- Click template names to view current configuration
### 3. Template Configuration
- **Normal templates**: Enter comma-separated values manually
- **Op templates**: Choose from BRAIN operators or enter manually
- **Data templates**: Enter data field specifications
### 4. Template Decoding
Two options for generating expressions:
- **full_iteration**: Generate all possible combinations
- Creates every possible combination using Cartesian product
- Shows first 999 results with search functionality for large datasets
- **random_iteration**: Generate random sample
- Specify number of random expressions (default: 10)
- Generates all combinations first, then randomly selects the specified number
- Useful for testing or when you only need a subset of possibilities
- Each template must have at least one configured value
- Results are displayed with copy and download options
## 💡 Example Expression
```
positive_sentiment = rank(<backfill_op/>(<positive_sentiment/>,<days/>));
negative_sentiment = rank(<backfill_op/>(<negative_sentiment/>, <days/>));
sentiment_difference = <compare_op/>(positive_sentiment, negative_sentiment);
<time_series_operator/>(sentiment_difference, <days/>)
```
### Example Configuration
- `<days/>``7, 14, 30`
- `<compare_op/>``ts_max, ts_min` (selected from BRAIN)
- `<time_series_operator/>``ts_delta, ts_mean` (selected from BRAIN)
### Generated Output
The decoder will generate all combinations:
```
ts_delta(ts_max(7), 7)
ts_delta(ts_max(14), 14)
ts_delta(ts_max(30), 30)
ts_delta(ts_min(7), 7)
ts_delta(ts_min(14), 14)
ts_delta(ts_min(30), 30)
ts_mean(ts_max(7), 7)
...and so on
```
## 🎯 Grammar Rules
- Use `/* */` for multi-line comments
- End statements with `;` (except the last line)
- No classes, objects, pointers, or functions allowed
- The last line is the Alpha expression for BRAIN simulator
## 🏗 Technical Architecture
### Flask Application Structure
```
BRAINProject/
├── app.py # Flask application with auto-dependency installation
├── templates/
│ └── index.html # Main web interface template
├── static/
│ ├── styles.css # Application styling
│ ├── script.js # Core application logic
│ ├── decoder.js # Template decoding algorithms
│ └── brain.js # BRAIN API integration module
├── requirements.txt # Python dependencies
├── mirror_config.txt # PyPI mirror configuration
├── run_app.bat # Windows startup script
├── run_app.sh # Unix/macOS startup script
├── setup_tsinghua.bat # Windows setup with Tsinghua mirror (for China)
├── setup_tsinghua.sh # Unix setup with Tsinghua mirror (for China)
├── brain_function_tester.ipynb # Jupyter notebook for testing
└── README.md # This file
```
### Key Components
- **Flask Backend**: Handles BRAIN API authentication and data fetching
- **Template Engine**: Jinja2 templates for dynamic content
- **Static Assets**: CSS and JavaScript served efficiently
- **CORS Handling**: Built-in cross-origin request support
- **Session Management**: Secure session handling for BRAIN authentication
### API Endpoints
- `GET /` - Main application page
- `POST /api/authenticate` - BRAIN authentication
- `GET /api/operators` - Fetch BRAIN operators with pagination
- `GET /api/datafields` - Fetch BRAIN data fields
- `GET /api/status` - Check authentication status
- `POST /api/logout` - Logout from BRAIN
## 🔧 Advanced Features
### Operator Fetching
- **Robust Pagination**: Handles BRAIN API pagination automatically
- **Category Support**: Fetches operators from all categories
- **Error Handling**: Multiple fallback strategies for reliable data loading
- **Debugging**: Built-in endpoints for API response analysis
### Template Processing
- **Real-time Detection**: Templates detected as you type
- **Visual Feedback**: Color-coded syntax highlighting
- **Status Indicators**: Clear visual cues for template configuration state
- **Validation**: Ensures all templates are configured before decoding
### User Experience
- **Professional UI**: Clean, modern interface design
- **Responsive Design**: Works on desktop and tablet devices
- **Keyboard Shortcuts**: Efficient template autocomplete
- **Export Options**: Copy individual results or download all
## 🐛 Troubleshooting
### Common Issues
1. **Flask server won't start**:
- Check Python version (3.7+ required)
- Install missing dependencies: `pip install -r requirements.txt`
2. **BRAIN connection fails**:
- Verify your BRAIN credentials
- Check internet connectivity
- Ensure no firewall blocking outbound HTTPS requests
3. **Templates not detecting**:
- Check template syntax: `<template_name/>`
- Refresh templates manually using the button
4. **No operators loading**:
- Verify BRAIN authentication is successful
- Check browser console for error messages
- Try logout and re-authentication
### Debug Mode
Run the application in debug mode for detailed error information:
```bash
python app.py
# Debug mode is enabled by default in the application
```
## 📈 Performance
- **Efficient Loading**: Operators loaded once during authentication
- **Pagination Handling**: Automatically fetches all data regardless of API pagination
- **Session Management**: Persistent sessions for better user experience
- **Optimized Frontend**: Minimal JavaScript dependencies for fast loading
## 🔐 Security
- **Session Security**: Flask secure sessions with configurable secret key
- **Credential Handling**: Credentials never stored, only used for API authentication
- **CORS Protection**: Properly configured cross-origin request handling
- **Input Validation**: Template syntax validation and grammar checking
## 🚧 Future Enhancements
- **Template Library**: Save and share template configurations
- **Advanced Filtering**: More sophisticated operator and data field filtering
- **Batch Operations**: Process multiple expressions simultaneously
- **API Integration**: Connect to additional financial data providers
- **Export Formats**: Support for JSON, CSV, and other output formats
---
**Note**: This Flask application replaces the previous HTML+proxy architecture, providing a more robust and integrated solution for BRAIN template decoding.