#!/usr/bin/env python3
"""
Fetch brand SVG logos from worldvectorlogo.com
Saves to /var/www/sneakerpicks/public/brands/{slug}.svg
Updates brands.logo_url in DB
"""
import requests
import psycopg2
import psycopg2.extras
import os
import re
import time
import sys
from urllib.parse import quote

DB_URL = "postgresql://supabase_admin:HzWrn2rMMxUYWO6OT5X0PScIVEvSjQMup9YSLIrkPQ@172.18.0.5:5432/sneakerpicks"
LOGO_DIR = "/var/www/sneakerpicks/public/brands"
BASE_URL = "https://worldvectorlogo.com"
# Alternative sources to try
ALT_SOURCES = [
    "https://cdn.worldvectorlogo.com/logos/{slug}.svg",
    "https://cdn.worldvectorlogo.com/logos/{slug}-1.svg",
    "https://cdn.worldvectorlogo.com/logos/{slug}-2.svg",
]

HEADERS = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
    "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
}

# Brand name → worldvectorlogo slug overrides
SLUG_OVERRIDES = {
    'nike': 'nike-4',
    'adidas': 'adidas-9',
    'adidas originals': 'adidas-trefoil',
    'new balance': 'new-balance-2',
    'puma': 'puma-6',
    'asics': 'asics',
    'jordan': 'air-jordan',
    'reebok': 'reebok-2019',
    'converse': 'converse-2',
    'vans': 'vans-1',
    'saucony': 'saucony',
    'hoka': 'hoka-one-one',
    'on running': 'on-running',
    'salomon': 'salomon-3',
    'fila': 'fila-1',
    'under armour': 'under-armour-1',
    'dr. martens': 'dr-martens',
    'tommy hilfiger': 'tommy-hilfiger-2',
    'calvin klein': 'calvin-klein-2',
    'hugo boss': 'hugo-boss',
    'lacoste': 'lacoste-1',
    'birkenstock': 'birkenstock',
    'the north face': 'the-north-face',
    'timberland': 'timberland-2',
    'skechers': 'skechers-1',
    'ugg': 'ugg-1',
    'crocs': 'crocs-1',
    'k-swiss': 'k-swiss-1',
    'le coq sportif': 'le-coq-sportif',
    'diadora': 'diadora-1',
    'mizuno': 'mizuno-2',
    'brooks': 'brooks-running',
    'filling pieces': 'filling-pieces',
    'golden goose': 'golden-goose',
    'alexander mcqueen': 'alexander-mcqueen',
    'versace': 'versace-3',
    'gucci': 'gucci-4',
    'balenciaga': 'balenciaga',
    'off-white': 'off-white-3',
    'stone island': 'stone-island',
    'y-3': 'y-3',
    'boss': 'boss-hugo-boss',
    'ellesse': 'ellesse',
    'umbro': 'umbro-1',
    'clarks': 'clarks',
    'ecco': 'ecco-2',
    'geox': 'geox',
    'camper': 'camper-1',
    'superga': 'superga',
    'fred perry': 'fred-perry-1',
    'levi\'s': 'levis-5',
    'g-star': 'g-star-raw',
    'diesel': 'diesel-3',
    'polo ralph lauren': 'polo-ralph-lauren',
    'ralph lauren': 'ralph-lauren-2',
    'michael kors': 'michael-kors',
    'coach': 'coach-2',
    'guess': 'guess-1',
}

# Minimum product count to bother fetching a logo
MIN_PRODUCTS = 3


def try_download_svg(url):
    """Try to download an SVG from a URL. Returns content or None."""
    try:
        r = requests.get(url, headers=HEADERS, timeout=15, allow_redirects=True)
        if r.status_code == 200 and ('svg' in r.headers.get('content-type', '') or r.text.strip().startswith('<?xml') or '<svg' in r.text[:500]):
            return r.text
    except:
        pass
    return None


def search_worldvectorlogo(brand_name):
    """Search worldvectorlogo.com for a brand and return the SVG URL."""
    # Try the search page
    search_url = f"{BASE_URL}/search/{quote(brand_name.lower())}"
    try:
        r = requests.get(search_url, headers=HEADERS, timeout=15)
        if r.status_code != 200:
            return None
        
        # Find SVG download links in the HTML
        # Pattern: /download/{slug} or cdn link
        matches = re.findall(r'href="(/logos/[^"]+)"', r.text)
        if matches:
            # Get the first logo page
            logo_page_url = f"{BASE_URL}{matches[0]}"
            r2 = requests.get(logo_page_url, headers=HEADERS, timeout=15)
            if r2.status_code == 200:
                # Find download link
                svg_matches = re.findall(r'href="(https://cdn\.worldvectorlogo\.com/logos/[^"]+\.svg)"', r2.text)
                if svg_matches:
                    return svg_matches[0]
        
        # Try to find direct CDN links in search results
        cdn_matches = re.findall(r'(https://cdn\.worldvectorlogo\.com/logos/[^"]+\.svg)', r.text)
        if cdn_matches:
            return cdn_matches[0]
            
    except Exception as e:
        pass
    return None


def fetch_logo(brand_name, brand_slug):
    """Try multiple strategies to get a brand logo SVG."""
    name_lower = brand_name.lower().strip()
    
    # 1. Check overrides
    if name_lower in SLUG_OVERRIDES:
        wvl_slug = SLUG_OVERRIDES[name_lower]
        url = f"https://cdn.worldvectorlogo.com/logos/{wvl_slug}.svg"
        svg = try_download_svg(url)
        if svg:
            return svg, url
    
    # 2. Try direct CDN with brand slug variations
    slugs_to_try = [
        brand_slug,
        brand_slug.replace('-', ' ').replace(' ', '-'),
        re.sub(r'-\d+$', '', brand_slug),  # remove trailing numbers
        f"{brand_slug}-1",
        f"{brand_slug}-2",
    ]
    
    for slug in slugs_to_try:
        url = f"https://cdn.worldvectorlogo.com/logos/{slug}.svg"
        svg = try_download_svg(url)
        if svg:
            return svg, url
    
    # 3. Search worldvectorlogo.com
    svg_url = search_worldvectorlogo(brand_name)
    if svg_url:
        svg = try_download_svg(svg_url)
        if svg:
            return svg, svg_url
    
    return None, None


def main():
    force = '--force' in sys.argv
    only_brand = None
    for arg in sys.argv[1:]:
        if arg.startswith('--brand='):
            only_brand = arg.split('=', 1)[1]
    
    print("=== SneakerPicks Brand Logo Fetcher ===\n")
    
    os.makedirs(LOGO_DIR, exist_ok=True)
    
    conn = psycopg2.connect(DB_URL)
    conn.autocommit = True
    cur = conn.cursor(cursor_factory=psycopg2.extras.DictCursor)
    
    # Get brands with product counts
    if only_brand:
        cur.execute("""
            SELECT b.id, b.name, b.slug, b.logo_url, COUNT(p.id) as product_count
            FROM brands b LEFT JOIN products p ON p.brand_id = b.id
            WHERE b.slug = %s OR b.name ILIKE %s
            GROUP BY b.id ORDER BY product_count DESC
        """, (only_brand, f"%{only_brand}%"))
    else:
        cur.execute("""
            SELECT b.id, b.name, b.slug, b.logo_url, COUNT(p.id) as product_count
            FROM brands b LEFT JOIN products p ON p.brand_id = b.id
            GROUP BY b.id
            HAVING COUNT(p.id) >= %s
            ORDER BY COUNT(p.id) DESC
        """, (MIN_PRODUCTS,))
    
    brands = cur.fetchall()
    print(f"Found {len(brands)} brands to process\n")
    
    fetched = 0
    skipped = 0
    failed = 0
    
    for brand in brands:
        brand_id = brand['id']
        name = brand['name']
        slug = brand['slug']
        existing_logo = brand['logo_url']
        count = brand['product_count']
        
        svg_path = os.path.join(LOGO_DIR, f"{slug}.svg")
        
        # Skip if already has logo and file exists (unless --force)
        if not force and existing_logo and os.path.exists(svg_path):
            skipped += 1
            continue
        
        print(f"  [{count:5d} products] {name} ({slug})...", end=" ", flush=True)
        
        svg_content, source_url = fetch_logo(name, slug)
        
        if svg_content:
            with open(svg_path, 'w', encoding='utf-8') as f:
                f.write(svg_content)
            
            logo_url = f"/brands/{slug}.svg"
            cur.execute("UPDATE brands SET logo_url = %s WHERE id = %s", (logo_url, brand_id))
            
            print(f"✓ saved ({len(svg_content)} bytes)")
            fetched += 1
        else:
            print("✗ not found")
            failed += 1
        
        # Rate limit
        time.sleep(0.5)
    
    print(f"\n{'='*40}")
    print(f"Fetched: {fetched}")
    print(f"Skipped (already have): {skipped}")
    print(f"Not found: {failed}")
    print(f"Total brands processed: {len(brands)}")
    
    # Show brands with logos
    cur.execute("SELECT COUNT(*) FROM brands WHERE logo_url IS NOT NULL")
    print(f"Brands with logos in DB: {cur.fetchone()[0]}")
    
    cur.close()
    conn.close()


if __name__ == "__main__":
    main()