import requests
import json
import base64
from concurrent.futures import ThreadPoolExecutor, as_completed
from collections import defaultdict
from urllib.parse import quote

# ================== 配置区 ==================
URL = "https://serverlist.piaservers.net/shadow_socks"
THREADS = 4                  # 建议改成 3~4，OpenWrt/LEDE 上并发太高容易出问题
TOTAL_REQUESTS = 500         # 可以改大到 800~1000
TIMEOUT = 12
# ===========================================

def fetch_once():
    try:
        headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
        }
        r = requests.get(URL, timeout=TIMEOUT, headers=headers)
        r.raise_for_status()
        text = r.text.strip()

        # 处理 "JSON\nbase64" 格式，取第一部分（JSON）
        if '\n' in text:
            json_text = text.split('\n', 1)[0].strip()
        else:
            json_text = text

        # 调试：打印前200字符，帮助排查
        # print("Raw start:", repr(json_text[:200]))

        data = json.loads(json_text)
        if isinstance(data, list):
            return data
        return None
    except Exception as e:
        # print("Request error:", type(e).__name__)  # 调试时可以取消注释
        return None

print(f"开始使用 {THREADS} 个线程采集 {TOTAL_REQUESTS} 次...")

servers = defaultdict(set)

with ThreadPoolExecutor(max_workers=THREADS) as executor:
    futures = [executor.submit(fetch_once) for _ in range(TOTAL_REQUESTS)]
    
    for i, future in enumerate(as_completed(futures), 1):
        data = future.result()
        if data:
            for item in data:
                if isinstance(item, dict):
                    region = str(item.get("region", "")).strip()
                    host = str(item.get("host", "")).strip()
                    if region and host:
                        servers[region].add(host)
        
        if i % 50 == 0 or i == TOTAL_REQUESTS:
            current = sum(len(hosts) for hosts in servers.values())
            print(f"已完成 {i}/{TOTAL_REQUESTS} 次请求... 当前发现节点数: {current}")

print("\n采集完成！")
total_nodes = sum(len(hosts) for hosts in servers.values())
print(f"共发现 {total_nodes} 个独特节点\n")

if total_nodes == 0:
    print("仍然采集不到节点。")
    print("可能原因：")
    print("1. 你的网络环境对这个域名做了特殊处理或被限流")
    print("2. 运行在 OpenWrt/LEDE 上，请求头或 TLS 有问题")
    print("建议：先用 curl 多测试几次，看是否每次都能正常返回 JSON")
    print("   curl -I https://serverlist.piaservers.net/shadow_socks")
else:
    # 输出表格
    print("地区                  Host数量    Hosts")
    print("-" * 85)
    all_nodes = []
    for region in sorted(servers.keys()):
        hosts = sorted(servers[region])
        print(f"{region:<20} {len(hosts):<8} {', '.join(hosts)}")
        for host in hosts:
            all_nodes.append((region, host))

    # 生成 ss:// 链接
    print(f"\n=== 生成 {len(all_nodes)} 个 ss:// 节点 ===")
    ss_links = []
    for region, host in all_nodes:
        auth = f"aes-128-gcm:shadowsocks@{host}:443"
        b64 = base64.urlsafe_b64encode(auth.encode()).decode().rstrip("=")
        name = f"PIA-{region}-{host.split('.')[-1]}"
        ss_url = f"ss://{b64}#{quote(name)}"
        ss_links.append(ss_url)
        print(ss_url)

    sub_url = "https://api.ss-sub.com/sub?target=ss&url=" + "|".join(ss_links)
    print(f"\n=== 一键订阅链接 ===")
    print(sub_url)

    # 保存到文件
    with open("pia_shadowsocks_nodes.txt", "w", encoding="utf-8") as f:
        f.write("\n".join(ss_links))
        f.write(f"\n\n订阅链接：\n{sub_url}")

    print("\n所有节点已保存到 pia_shadowsocks_nodes.txt 文件中！")