#!/usr/bin/env python3
"""
IMPORTADOR COMPLETO DEL CATÁLOGO GONHER 2022
- Crea vehículos faltantes
- Crea partes de filtros Gonher
- Crea referencias cruzadas con otras marcas (AC Delco, Fram, etc.)
- Crea fitments (vincula partes a vehículos)
"""

import sqlite3
import re
import pypdf
from pathlib import Path
from collections import defaultdict

# Paths
DB_PATH = Path(__file__).parent.parent / 'vehicle_database.db'
PDF_PATH = '/tmp/filtros_catalog.pdf'

# Filter type configuration
FILTER_TYPES = {
    'ACEITE': {'category': 'Engine', 'group': 'Oil Filters', 'prefix': 'Oil Filter', 'prefix_es': 'Filtro de Aceite'},
    'SINTÉTICO': {'category': 'Engine', 'group': 'Oil Filters', 'prefix': 'Synthetic Oil Filter', 'prefix_es': 'Filtro de Aceite Sintético'},
    'AIRE': {'category': 'Engine', 'group': 'Air Filters', 'prefix': 'Air Filter', 'prefix_es': 'Filtro de Aire'},
    'COMB.': {'category': 'Fuel & Air', 'group': 'Fuel Filters', 'prefix': 'Fuel Filter', 'prefix_es': 'Filtro de Combustible'},
    'CABINA': {'category': 'Heat & Air Conditioning', 'group': 'Cabin Air Filters', 'prefix': 'Cabin Air Filter', 'prefix_es': 'Filtro de Cabina'}
}

# Known brands in catalog
CATALOG_BRANDS = [
    'ACURA', 'ALFA ROMEO', 'AUDI', 'BMW', 'BUICK', 'CADILLAC',
    'CHEVROLET', 'CHRYSLER', 'DODGE', 'FIAT', 'FORD', 'GMC',
    'HONDA', 'HYUNDAI', 'INFINITI', 'JAGUAR', 'JEEP', 'KIA',
    'LEXUS', 'LINCOLN', 'MAZDA', 'MERCEDES BENZ', 'MERCURY',
    'MINI', 'MITSUBISHI', 'NISSAN', 'PEUGEOT', 'PONTIAC',
    'PORSCHE', 'RAM', 'RENAULT', 'SEAT', 'SMART', 'SUBARU',
    'SUZUKI', 'TOYOTA', 'VOLKSWAGEN', 'VOLVO'
]

# Cross-reference brands
XREF_BRANDS = ['AC DELCO', 'FRAM', 'INTERFIL', 'MANN', 'MOTORCRAFT']

def get_db():
    conn = sqlite3.connect(DB_PATH)
    conn.row_factory = sqlite3.Row
    return conn

def ensure_manufacturer(cursor, name, type_='aftermarket', quality='standard', country=None):
    """Create manufacturer if not exists"""
    cursor.execute("SELECT id FROM manufacturers WHERE UPPER(name) = UPPER(?)", (name,))
    row = cursor.fetchone()
    if row:
        return row['id']

    cursor.execute("""
        INSERT INTO manufacturers (name, type, quality_tier, country)
        VALUES (?, ?, ?, ?)
    """, (name, type_, quality, country))
    return cursor.lastrowid

def ensure_brand(cursor, name):
    """Create brand if not exists"""
    cursor.execute("SELECT id FROM brands WHERE UPPER(name) = UPPER(?)", (name,))
    row = cursor.fetchone()
    if row:
        return row['id']

    cursor.execute("INSERT INTO brands (name) VALUES (?)", (name,))
    return cursor.lastrowid

def ensure_model(cursor, brand_id, name):
    """Create model if not exists"""
    cursor.execute("""
        SELECT id FROM models
        WHERE brand_id = ? AND UPPER(name) = UPPER(?)
    """, (brand_id, name))
    row = cursor.fetchone()
    if row:
        return row['id']

    cursor.execute("INSERT INTO models (brand_id, name) VALUES (?, ?)", (brand_id, name))
    return cursor.lastrowid

def ensure_year(cursor, year):
    """Create year if not exists"""
    cursor.execute("SELECT id FROM years WHERE year = ?", (year,))
    row = cursor.fetchone()
    if row:
        return row['id']

    cursor.execute("INSERT INTO years (year) VALUES (?)", (year,))
    return cursor.lastrowid

def ensure_engine(cursor, name):
    """Create engine if not exists"""
    cursor.execute("SELECT id FROM engines WHERE name = ?", (name,))
    row = cursor.fetchone()
    if row:
        return row['id']

    # Parse engine details from name
    displacement = None
    cylinders = None
    fuel_type = 'gasoline'  # lowercase to match DB constraint

    # Parse displacement and cylinders from patterns like "L4-2.0L" or "V6-3.5L"
    match = re.match(r'([LV])(\d+)[-]?([\d.]+)L?', name)
    if match:
        engine_type = match.group(1)  # L or V
        cylinders = int(match.group(2))
        displacement = int(float(match.group(3)) * 1000)

    if 'DIESEL' in name.upper() or 'DIÉSEL' in name.upper():
        fuel_type = 'diesel'
    elif 'ELECTRIC' in name.upper():
        fuel_type = 'electric'
    elif 'HYBRID' in name.upper():
        fuel_type = 'hybrid'
    # Note: 'TURBO' is not a fuel type, it's a modifier - default to gasoline

    cursor.execute("""
        INSERT INTO engines (name, displacement_cc, cylinders, fuel_type)
        VALUES (?, ?, ?, ?)
    """, (name, displacement, cylinders, fuel_type))
    return cursor.lastrowid

def ensure_mye(cursor, model_id, year_id, engine_id):
    """Create model_year_engine if not exists"""
    cursor.execute("""
        SELECT id FROM model_year_engine
        WHERE model_id = ? AND year_id = ? AND engine_id = ?
    """, (model_id, year_id, engine_id))
    row = cursor.fetchone()
    if row:
        return row['id']

    cursor.execute("""
        INSERT INTO model_year_engine (model_id, year_id, engine_id)
        VALUES (?, ?, ?)
    """, (model_id, year_id, engine_id))
    return cursor.lastrowid

def get_or_create_group(cursor, category_name, group_name):
    """Get or create part group"""
    cursor.execute("SELECT id FROM part_categories WHERE name = ?", (category_name,))
    cat_row = cursor.fetchone()
    if not cat_row:
        return None

    category_id = cat_row['id']

    cursor.execute("""
        SELECT id FROM part_groups WHERE category_id = ? AND name = ?
    """, (category_id, group_name))
    group_row = cursor.fetchone()

    if group_row:
        return group_row['id']

    cursor.execute("""
        INSERT INTO part_groups (category_id, name, name_es)
        VALUES (?, ?, ?)
    """, (category_id, group_name, group_name))
    return cursor.lastrowid

def parse_year_range(year_str):
    """Parse year range into list of years"""
    year_str = year_str.strip()

    if re.match(r'^\d{4}$', year_str):
        return [int(year_str)]

    match = re.match(r'(\d{4})\s*[-–]\s*(\d{4})', year_str)
    if match:
        start, end = int(match.group(1)), int(match.group(2))
        if start > end:
            start, end = end, start
        return list(range(end, start + 1))

    return []

def classify_filter(part_number):
    """Classify filter type by part number prefix"""
    part_number = part_number.upper()
    if part_number.startswith('GP-') or part_number.startswith('GPS-'):
        if part_number.startswith('GPS-'):
            return 'SINTÉTICO'
        return 'ACEITE'
    elif part_number.startswith('GA-') and not part_number.startswith('GAC-') and not part_number.startswith('GAVW-'):
        return 'AIRE'
    elif part_number.startswith('GG-'):
        return 'COMB.'
    elif part_number.startswith('GAC-'):
        return 'CABINA'
    elif part_number.startswith('G-'):
        return 'ACEITE'  # Generic oil filter
    return None

def extract_vehicle_entries(pdf):
    """Extract all vehicle entries from catalog"""
    entries = []
    current_brand = None
    current_model = None

    for page in pdf.pages:
        text = page.extract_text()
        if not text:
            continue

        for line in text.split('\n'):
            line = line.strip()
            if not line:
                continue

            # Skip header/footer
            if 'AÑO' in line and 'MOTOR' in line:
                continue
            if 'Los filtros Gonher' in line:
                continue

            # Brand detection
            if line in CATALOG_BRANDS:
                current_brand = line
                current_model = None
                continue

            # Handle (Continúa) lines
            if '(Continúa)' in line:
                match = re.match(r'^([A-Z][A-Z0-9\s\-]+)\s*\(Continúa\)', line)
                if match:
                    potential = match.group(1).strip()
                    if potential in CATALOG_BRANDS:
                        current_brand = potential
                    elif current_brand:
                        current_model = potential
                continue

            # Model detection
            if current_brand:
                if re.match(r'^[A-Z][A-Z0-9\s\-/]+$', line) and not re.match(r'^\d{4}', line):
                    if line not in ['ACEITE', 'AIRE', 'COMB', 'CABINA', 'SINTÉTICO', 'AÑO', 'MOTOR']:
                        if not re.match(r'^G[APCS]?[-]?\d', line):  # Not a part number
                            current_model = line
                            continue

            # Data line with year
            if current_brand and current_model:
                match = re.match(r'^(\d{4}(?:\s*[-–]\s*\d{4})?)\s+(.+)$', line)
                if match:
                    year_str = match.group(1)
                    rest = match.group(2)

                    # Extract motor
                    motor_match = re.match(r'^([LV]\d+[-][\d.]+L(?:\s+(?:Turbo|TURBOCHARGED|diésel|ELECTRIC|HYBRID))?)\s*(.*)$', rest, re.IGNORECASE)
                    if motor_match:
                        motor = motor_match.group(1).strip()
                        filters_str = motor_match.group(2).strip()

                        # Parse filter part numbers
                        filter_parts = re.findall(r'G[A-Z]*[-]?[\dA-Z]+(?:\(\d+\))?', filters_str)

                        years = parse_year_range(year_str)

                        if years:
                            for year in years:
                                entry = {
                                    'brand': current_brand,
                                    'model': current_model,
                                    'year': year,
                                    'motor': motor,
                                    'filters': {}
                                }

                                for fp in filter_parts:
                                    fp_clean = re.sub(r'\(\d+\)', '', fp)
                                    filter_type = classify_filter(fp_clean)
                                    if filter_type:
                                        entry['filters'][filter_type] = fp_clean

                                if entry['filters']:
                                    entries.append(entry)

    return entries

def extract_cross_references(pdf):
    """Extract cross-reference data from catalog"""
    xrefs = []
    current_brand = None

    # Cross-references are typically in pages 117+
    for i in range(117, len(pdf.pages)):
        text = pdf.pages[i].extract_text()
        if not text:
            continue

        for line in text.split('\n'):
            line = line.strip()

            # Brand header
            if line in XREF_BRANDS:
                current_brand = line
                continue

            # Cross-reference line
            if current_brand:
                match = re.match(r'^([A-Z0-9\-/]+)\s+(G[A-Z]*[-]?\d+[A-Z]*)$', line)
                if match:
                    xrefs.append({
                        'brand': current_brand,
                        'part_number': match.group(1),
                        'gonher_part': match.group(2)
                    })

    return xrefs

def main():
    print("=" * 70)
    print("IMPORTADOR COMPLETO - CATÁLOGO GONHER 2022")
    print("=" * 70)

    # Read PDF
    print(f"\n[1/7] Leyendo PDF: {PDF_PATH}")
    pdf = pypdf.PdfReader(PDF_PATH)
    print(f"      Total páginas: {len(pdf.pages)}")

    # Extract data
    print("\n[2/7] Extrayendo datos del catálogo...")
    vehicle_entries = extract_vehicle_entries(pdf)
    cross_refs = extract_cross_references(pdf)
    print(f"      Entradas de vehículos: {len(vehicle_entries)}")
    print(f"      Referencias cruzadas: {len(cross_refs)}")

    # Get unique filters
    unique_filters = {}
    for entry in vehicle_entries:
        for filter_type, part_num in entry['filters'].items():
            if part_num not in unique_filters:
                unique_filters[part_num] = filter_type
    print(f"      Filtros únicos: {len(unique_filters)}")

    # Connect to database
    conn = get_db()
    cursor = conn.cursor()

    # Create manufacturers
    print("\n[3/7] Creando fabricantes...")
    manufacturers = {
        'Gonher': ensure_manufacturer(cursor, 'Gonher', 'aftermarket', 'standard', 'Mexico'),
        'AC Delco': ensure_manufacturer(cursor, 'AC Delco', 'oem', 'oem', 'USA'),
        'Fram': ensure_manufacturer(cursor, 'Fram', 'aftermarket', 'standard', 'USA'),
        'Interfil': ensure_manufacturer(cursor, 'Interfil', 'aftermarket', 'economy', 'Mexico'),
        'Mann': ensure_manufacturer(cursor, 'Mann', 'aftermarket', 'premium', 'Germany'),
        'Motorcraft': ensure_manufacturer(cursor, 'Motorcraft', 'oem', 'oem', 'USA'),
    }
    print(f"      Fabricantes: {list(manufacturers.keys())}")

    # Create vehicles
    print("\n[4/7] Creando vehículos faltantes...")
    vehicles_created = 0
    mye_cache = {}

    for entry in vehicle_entries:
        cache_key = (entry['brand'], entry['model'], entry['year'], entry['motor'])
        if cache_key in mye_cache:
            continue

        # Check if vehicle exists
        cursor.execute("""
            SELECT mye.id FROM model_year_engine mye
            JOIN models m ON mye.model_id = m.id
            JOIN brands b ON m.brand_id = b.id
            JOIN years y ON mye.year_id = y.id
            JOIN engines e ON mye.engine_id = e.id
            WHERE UPPER(b.name) = UPPER(?) AND UPPER(m.name) = UPPER(?)
                  AND y.year = ? AND e.name = ?
            LIMIT 1
        """, (entry['brand'], entry['model'], entry['year'], entry['motor']))

        existing = cursor.fetchone()
        if existing:
            mye_cache[cache_key] = existing['id']
        else:
            # Create vehicle
            brand_id = ensure_brand(cursor, entry['brand'])
            model_id = ensure_model(cursor, brand_id, entry['model'])
            year_id = ensure_year(cursor, entry['year'])
            engine_id = ensure_engine(cursor, entry['motor'])
            mye_id = ensure_mye(cursor, model_id, year_id, engine_id)
            mye_cache[cache_key] = mye_id
            vehicles_created += 1

    print(f"      Vehículos creados: {vehicles_created}")

    # Create filter parts
    print("\n[5/7] Creando partes de filtros...")
    filter_parts = {}
    parts_created = 0

    for part_num, filter_type in unique_filters.items():
        config = FILTER_TYPES.get(filter_type)
        if not config:
            continue

        group_id = get_or_create_group(cursor, config['category'], config['group'])
        if not group_id:
            continue

        # Check if part exists
        cursor.execute("SELECT id FROM parts WHERE oem_part_number = ?", (part_num,))
        existing = cursor.fetchone()

        if existing:
            filter_parts[part_num] = existing['id']
        else:
            name = f"{config['prefix']} {part_num}"
            name_es = f"{config['prefix_es']} {part_num}"

            cursor.execute("""
                INSERT INTO parts (oem_part_number, name, name_es, group_id, description)
                VALUES (?, ?, ?, ?, ?)
            """, (part_num, name, name_es, group_id, f"Gonher {config['prefix']}"))

            filter_parts[part_num] = cursor.lastrowid
            parts_created += 1

    print(f"      Partes creadas: {parts_created}")

    # Create fitments
    print("\n[6/7] Creando fitments (vehículo-parte)...")
    fitments_created = 0

    for entry in vehicle_entries:
        cache_key = (entry['brand'], entry['model'], entry['year'], entry['motor'])
        mye_id = mye_cache.get(cache_key)

        if not mye_id:
            continue

        for filter_type, part_num in entry['filters'].items():
            part_id = filter_parts.get(part_num)
            if not part_id:
                continue

            # Check if fitment exists
            cursor.execute("""
                SELECT id FROM vehicle_parts
                WHERE model_year_engine_id = ? AND part_id = ?
            """, (mye_id, part_id))

            if not cursor.fetchone():
                cursor.execute("""
                    INSERT INTO vehicle_parts (model_year_engine_id, part_id, quantity_required, fitment_notes)
                    VALUES (?, ?, 1, ?)
                """, (mye_id, part_id, f"Catálogo Gonher 2022 - {filter_type}"))
                fitments_created += 1

    print(f"      Fitments creados: {fitments_created}")

    # Create cross-references
    print("\n[7/7] Creando referencias cruzadas...")
    xrefs_created = 0

    for xref in cross_refs:
        gonher_part_id = filter_parts.get(xref['gonher_part'])
        if not gonher_part_id:
            # Part might not exist yet, try to find by OEM number
            cursor.execute("SELECT id FROM parts WHERE oem_part_number = ?", (xref['gonher_part'],))
            row = cursor.fetchone()
            if row:
                gonher_part_id = row['id']
            else:
                continue

        # Check if cross-reference exists
        cursor.execute("""
            SELECT id FROM part_cross_references
            WHERE part_id = ? AND cross_reference_number = ?
        """, (gonher_part_id, xref['part_number']))

        if not cursor.fetchone():
            # Map brand to reference type
            ref_type = 'interchange'
            if xref['brand'] in ['AC DELCO', 'MOTORCRAFT']:
                ref_type = 'oem_alternate'

            cursor.execute("""
                INSERT INTO part_cross_references (part_id, cross_reference_number, reference_type)
                VALUES (?, ?, ?)
            """, (gonher_part_id, xref['part_number'], ref_type))
            xrefs_created += 1

    print(f"      Referencias cruzadas creadas: {xrefs_created}")

    # Commit
    conn.commit()
    conn.close()

    print("\n" + "=" * 70)
    print("IMPORTACIÓN COMPLETADA")
    print("=" * 70)
    print(f"""
RESUMEN:
  - Vehículos creados:     {vehicles_created:,}
  - Partes creadas:        {parts_created:,}
  - Fitments creados:      {fitments_created:,}
  - Cross-refs creadas:    {xrefs_created:,}
  - Fabricantes:           {len(manufacturers)}
""")

if __name__ == '__main__':
    main()