"""Download real dish images from 下厨房 for all 69 dishes."""
import json, os, re, time, random, requests

DIR = os.path.dirname(os.path.abspath(__file__))
IMAGES = os.path.join(DIR, 'images')
os.makedirs(IMAGES, exist_ok=True)

with open(os.path.join(DIR, 'dishes.json'), 'r', encoding='utf-8') as f:
    dishes = json.load(f)

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',
}

done = 0
fail = 0
none = 0
total = len(dishes)

for d in dishes:
    name = d['菜名']
    dish_id = d['id']
    fname = f'dish_{dish_id}.jpg'
    fpath = os.path.join(IMAGES, fname)

    # Skip if already downloaded
    if os.path.exists(fpath):
        done += 1
        continue

    print(f'[{done+fail+none+1}/{total}] {name} ...', end=' ', flush=True)

    try:
        resp = requests.get('https://www.xiachufang.com/search/',
                          params={'keyword': name},
                          headers=headers, timeout=15)
        imgs = re.findall(r'https?://i\d\.chuimg\.com/[a-f0-9]+_\d+w_\d+h\.jpg', resp.text)
        if not imgs:
            # Try PNG too
            imgs = re.findall(r'https?://i\d\.chuimg\.com/[a-f0-9]+_\d+w_\d+h\.(?:jpg|png|webp)', resp.text)

        if imgs:
            img_url = imgs[0]
            img_resp = requests.get(img_url, headers=headers, timeout=15)
            if img_resp.status_code == 200 and len(img_resp.content) > 5000:
                with open(fpath, 'wb') as f:
                    f.write(img_resp.content)
                print(f'OK ({len(img_resp.content)//1024}KB)')
                done += 1
            else:
                print(f'SKIP (HTTP {img_resp.status_code}, {len(img_resp.content)}B)')
                fail += 1
        else:
            print('NONE')
            none += 1
    except Exception as e:
        print(f'ERR: {e}')
        fail += 1

    time.sleep(1.5 + random.random() * 2)

print(f'\n=== 完成 ===')
print(f'成功: {done}  失败: {fail}  未找到: {none}  总计: {total}')