by-agent-for-agent/scripts/twitter_dl_v2.py

# /// script
# requires-python = ">=3.12"
# dependencies = [
#     "httpx",
# ]
# ///
import httpx
import re
import os
import sys

def download_twitter_photo_nitter(url, output_dir="downloads"):
    if not os.path.exists(output_dir):
        os.makedirs(output_dir)

    # 提取推文 ID
    match = re.search(r'status/(\d+)', url)
    if not match:
        print("❌ 无法提取推文 ID")
        return
    tweet_id = match.group(1)

    # 使用 Nitter 镜像站绕过限制（这是目前 Docker 环境最稳的方法）
    # Nitter 实例列表可以替换，如果这个挂了可以换另一个
    nitter_instances = ["https://nitter.net", "https://nitter.cz", "https://nitter.moomoo.me"]
    
    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"
    }

    success = False
    for instance in nitter_instances:
        try:
            nitter_url = f"{instance}/i/status/{tweet_id}"
            print(f"🕵️ 尝试通过镜像站抓取: {nitter_url}")
            
            with httpx.Client(headers=headers, follow_redirects=True, timeout=20.0) as client:
                response = client.get(nitter_url)
                if response.status_code != 200:
                    continue
                
                # 寻找图片链接 (Nitter 的图片通常是 /pic/media%2F...)
                # 示例正则匹配
                img_paths = re.findall(r'/pic/media%2F[^"]+', response.text)
                if not img_paths:
                    # 尝试另一种匹配
                    img_paths = re.findall(r'/pic/orig/media%2F[^"]+', response.text)

                if img_paths:
                    for i, path in enumerate(set(img_paths)):
                        full_img_url = f"{instance}{path}"
                        print(f"📥 发现图片: {full_img_url}")
                        
                        img_res = client.get(full_img_url)
                        if img_res.status_code == 200:
                            save_path = os.path.join(output_dir, f"tweet_{tweet_id}_{i}.jpg")
                            with open(save_path, "wb") as f:
                                f.write(img_res.content)
                            print(f"✅ 保存成功: {save_path}")
                            success = True
                    
                    if success: break # 下载到就停
        except Exception as e:
            print(f"⚠️ 尝试实例 {instance} 时出错: {e}")

    if not success:
        print("❌ 所有尝试均失败。可能镜像站也无法访问该推文，或该推文是私密的。")

if __name__ == "__main__":
    if len(sys.argv) < 2:
         print("用法: uv run scripts/twitter_dl_v2.py <url>")
    else:
        download_twitter_photo_nitter(sys.argv[1])