import re
import json

input_file = 'cologne_duesseldorf_data/duesseldorf_raw.txt'
output_json = 'cologne_duesseldorf_data/duesseldorf_targets.json'

def parse_targets():
    with open(input_file, 'r', encoding='utf-8') as f:
        lines = f.readlines()
    
    targets = []
    # Pattern: • {Innung}/OM: {Name}
    # Example: • Augenoptiker-Innung Düssel-Rhein-Ruhr/OM: Jens Schulz
    
    pattern = re.compile(r'•\s*(.*?)/OM:\s*(.*)')
    
    for line in lines:
        line = line.strip()
        match = pattern.match(line)
        if match:
            innung = match.group(1).strip()
            name = match.group(2).strip()
            targets.append({
                "query": f"{innung} Düsseldorf Kontakt Email",
                "innung": innung,
                "person": name
            })
            
    print(f"Found {len(targets)} targets.")
    
    # Prioritize "Bau", "Elektro", "Sanitär", "Metall", "Dach"
    priority_keywords = ["Bau", "Elektro", "Sanitär", "Metall", "Dach", "Tischler"]
    sorted_targets = sorted(targets, key=lambda x: any(k in x['innung'] for k in priority_keywords), reverse=True)
    
    with open(output_json, 'w', encoding='utf-8') as f:
        json.dump(sorted_targets, f, indent=2)
        
    for t in sorted_targets[:5]:
        print(f"Target: {t['innung']} ({t['person']})")

if __name__ == "__main__":
    parse_targets()