import json, os, re, urllib.request, time

base = os.path.dirname(os.path.abspath(__file__))
out_root = os.path.join(base, "products")
os.makedirs(out_root, exist_ok=True)

def slugify(name):
    s = re.sub(r'[^a-zA-Z0-9]+', '-', name).strip('-').lower()
    return s[:60]

catalogue = []
with open(os.path.join(base, "products.jsonl")) as f:
    for line in f:
        line = line.strip()
        if not line:
            continue
        catalogue.append(json.loads(line))

for i, p in enumerate(catalogue, 1):
    dirname = f"{i:02d}-{slugify(p['brand'])}-{slugify(p['name'])}"
    pdir = os.path.join(out_root, dirname)
    os.makedirs(pdir, exist_ok=True)
    with open(os.path.join(pdir, "info.json"), "w") as f:
        json.dump({"name": p["name"], "brand": p["brand"], "price_myr": p["price"], "source_code": p["code"]}, f, indent=2)
    for j, url in enumerate(p["imgs"], 1):
        dest = os.path.join(pdir, f"image-{j}.jpg")
        try:
            req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0"})
            with urllib.request.urlopen(req, timeout=15) as r, open(dest, "wb") as out:
                out.write(r.read())
        except Exception as e:
            print(f"FAIL {url}: {e}")
    print(f"[{i}/{len(catalogue)}] {p['brand']} - {p['name']} -> {len(p['imgs'])} imgs")