unity_functions.py

import json
import re
from typing import Dict, Any

def extract_manual_data(content: str) -> Dict[str, Any]:
    """Manually extract product data when JSON parsing fails"""
    extracted_data = {}
    
    # Common patterns to extract information
    patterns = {
        'material_composition': r'material[s]?[:\-\s]*(.*?)(?:\n|$)',
        'manufacturing_location': r'manufactur[ing]*[:\-\s]*(?:in\s+)?(.*?)(?:\n|$)',
        'product_weight': r'weight[:\-\s]*(\d+(?:\.\d+)?)\s*(kg|g|pounds?|lbs?)',
        'transport_distance': r'distance[:\-\s]*(\d+(?:\.\d+)?)\s*(km|miles?)',
        'packaging_type': r'packag[ing]*[:\-\s]*(.*?)(?:\n|$)',
        'energy_usage': r'energy[:\-\s]*(\d+(?:\.\d+)?)\s*(kwh|watts?|w)',
    }
    
    content_lower = content.lower()
    
    for key, pattern in patterns.items():
        match = re.search(pattern, content_lower, re.IGNORECASE)
        if match:
            if 'weight' in key or 'distance' in key or 'energy' in key:
                # Extract numeric value and convert to standard units
                value = float(match.group(1))
                unit = match.group(2).lower()
                
                # Convert to standard units
                if key == 'product_weight':
                    if unit in ['g', 'grams']:
                        value = value / 1000  # Convert to kg
                    elif unit in ['pounds', 'lbs', 'lb']:
                        value = value * 0.453592  # Convert to kg
                    extracted_data[key] = value
                    
                elif key == 'transport_distance':
                    if unit in ['miles', 'mile']:
                        value = value * 1.60934  # Convert to km
                    extracted_data[key] = value
                    
                elif key == 'energy_usage':
                    if unit in ['watts', 'w']:
                        value = value / 1000  # Convert to kWh (assuming 1 hour usage)
                    extracted_data[key] = value
            else:
                extracted_data[key] = match.group(1).strip()
    
    # Set default values if not found
    if 'product_weight' not in extracted_data:
        extracted_data['product_weight'] = 0.5  # Default 0.5 kg
    
    if 'transport_distance' not in extracted_data:
        extracted_data['transport_distance'] = 1000  # Default 1000 km
    
    return extracted_data


def parse_extraction_result(content):
    """Extract structured data from LLM response"""
    try:
        # Look for JSON in the response
        json_match = re.search(r'\{.*\}', content, re.DOTALL)
        if json_match:
            return json.loads(json_match.group())
    except:
        pass
    
    # Fallback to manual parsing
    return extract_manual_data(content)

def parse_scoring_result(content):
    """Extract score and recommendations from LLM response"""
    score_match = re.search(r'(\d+)', content)
    score = int(score_match.group(1)) if score_match else 50
    
    # Extract recommendations
    recommendations = []
    lines = content.split('\n')
    for line in lines:
        if line.strip().startswith('-') or line.strip().startswith('•'):
            recommendations.append(line.strip()[1:].strip())
    
    return {
        "score": score,
        "recommendations": recommendations
    }