import re
from db import insert_lead

def parse_tradingtwins_email(body):
    data = {}
    
    # Simple regex extraction based on the email format
    patterns = {
        'id': r'Lead-ID:\s*(\d+)',
        'company': r'Firma:\s*(.+)',
        'contact_first': r'Vorname:\s*(.+)',
        'contact_last': r'Nachname:\s*(.+)',
        'email': r'E-Mail:\s*([^\s<]+)',
        'phone': r'Rufnummer:\s*([^\n]+)',
        'area': r'Reinigungs-Flche:\s*([^\n]+)',
        'purpose': r'Einsatzzweck:\s*([^\n]+)'
    }
    
    for key, pattern in patterns.items():
        match = re.search(pattern, body)
        if match:
            data[key] = match.group(1).strip()
            
    # Combine names
    if 'contact_first' in data and 'contact_last' in data:
        data['contact'] = f"{data['contact_first']} {data['contact_last']}"
    
    data['raw_body'] = body
    return data

def parse_roboplanet_form(body):
    """
    Parses the Roboplanet website contact form (HTML format).
    Example: <b>Vorname:</b> BÄKO <br><b>Nachname:</b> eG <br><b>Email:</b> Alexander.Grau@baeko-hr.de ...
    """
    data = {}
    
    # Helper to strip HTML tags if needed, but we'll use regex on the content
    patterns = {
        'contact_first': r'Vorname:</b>\s*(.*?)\s*<br>',
        'contact_last': r'Nachname:</b>\s*(.*?)\s*<br>',
        'email': r'Email:</b>\s*(.*?)\s*<br>',
        'phone': r'Telefon:</b>\s*(.*?)\s*<br>',
        'company': r'Firma:</b>\s*(.*?)\s*<br>',
        'zip': r'PLZ:</b>\s*(.*?)\s*<br>',
        'message': r'Nachricht:</b>\s*(.*?)\s*(?:<br>|--|$)'
    }
    
    for key, pattern in patterns.items():
        # Use re.DOTALL for message if it spans lines, but usually it's one block
        match = re.search(pattern, body, re.IGNORECASE | re.DOTALL)
        if match:
            # Clean HTML tags from the captured value if any
            val = re.sub(r'<.*?>', '', match.group(1)).strip()
            data[key] = val
            
    # Combine names
    if 'contact_first' in data and 'contact_last' in data:
        data['contact'] = f"{data['contact_first']} {data['contact_last']}"
    
    # For Roboplanet forms, we use the timestamp as ID or a hash if missing
    data['raw_body'] = body
    return data

def ingest_mock_leads():
    # Mock data from the session context
    leads = [
        {
            'id': '2397256',
            'company': 'pronorm Einbauküchen GmbH',
            'contact': 'Jakob Funk',
            'email': 'jakob.funk@pronorm.de',
            'phone': '+49 5733 979175',
            'raw_body': """
            Lead-ID: 2397256
            Firma: pronorm Einbauküchen GmbH
            Vorname: Jakob
            Nachname: Funk
            Reinigungs-Flche: 1.001 - 10.000 qm
            Einsatzzweck: Reinigung von Böden
            """
        },
        {
            'id': '2414364',
            'company': 'Quad & Rollershop Schwabmünchen GmbH',
            'contact': 'Manfred Bihler',
            'email': 'Rollershop.Schwabmuenchen@web.de',
            'phone': '+49 8232 905246',
            'raw_body': """
            Lead-ID: 2414364
            Firma: Quad & Rollershop Schwabmünchen GmbH
            Vorname: Manfred
            Nachname: Bihler
            Reinigungs-Flche: 301 - 1.000 qm
            Einsatzzweck: Reinigung von Böden
            """
        }
    ]
    
    count = 0
    for lead in leads:
        if insert_lead(lead):
            count += 1
    return count

if __name__ == "__main__":
    from db import init_db
    init_db()
    print(f"Ingested {ingest_mock_leads()} new leads.")