start.sh

#!/bin/bash

# Start script for the lightweight LLM service
# This script handles environment setup and service startup

set -e

echo "🚀 Starting Lightweight LLM with MCP Integration..."

# Print configuration
echo "📋 Configuration:"
echo "  Model: ${MODEL_NAME:-microsoft/DialoGPT-small}"
echo "  Port: ${PORT:-7860}"
echo "  MCP Server: ${MCP_SERVER_URL:-Not configured}"
echo "  Max Tokens: ${MAX_NEW_TOKENS:-256}"

# Health check function
health_check() {
    local max_attempts=30
    local attempt=1
    
    echo "🔍 Waiting for service to be ready..."
    
    while [ $attempt -le $max_attempts ]; do
        if curl -f http://localhost:${PORT:-7860}/health >/dev/null 2>&1; then
            echo "✅ Service is ready!"
            return 0
        fi
        
        echo "⏳ Attempt $attempt/$max_attempts - waiting 2 seconds..."
        sleep 2
        attempt=$((attempt + 1))
    done
    
    echo "❌ Service failed to start within expected time"
    return 1
}

# Start the application in the background
echo "🔄 Starting FastAPI application..."
python app.py &
APP_PID=$!

# Wait for the service to be ready
if health_check; then
    echo "🎉 Lightweight LLM service is running successfully!"
    echo "📡 API Documentation: http://localhost:${PORT:-7860}/docs"
    echo "🏥 Health Check: http://localhost:${PORT:-7860}/health"
    echo "ℹ️  Service Info: http://localhost:${PORT:-7860}/info"
    
    # Keep the script running
    wait $APP_PID
else
    echo "💥 Failed to start service"
    kill $APP_PID 2>/dev/null || true
    exit 1
fi