import requests
import re
import demjson3 as demjson
import json
import hashlib
import sys
import os
from pathlib import Path

# 创建全局 session 并设置浏览器 UA
session = requests.Session()
COMMON_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",
}
session.headers.update(COMMON_HEADERS)
session.cookies.set("visited", "1")  # 提高拟真度

# 下载伪 JSON 文本
def fetch_raw_json():
    url = "http://www.xn--sss604efuw.art/jm/jiemi.php?url=http%3A%2F%2Fwww.%E9%A5%AD%E5%A4%AA%E7%A1%AC.art%2Ftv"
    resp = session.get(url, timeout=30, allow_redirects=True)
    resp.encoding = 'utf-8'
    return resp.text

# 读取本地 JSON 文件
def read_local_json(file_path):
    """读取本地JSON文件，支持带注释的JSON和图片中的base64数据"""
    with open(file_path, 'rb') as f:
        raw_content = f.read()

    # 检查是否为图片文件
    image_headers = [
        b'\xff\xd8\xff\xe0',  # JPEG
        b'\xff\xd8\xff\xe1',  # JPEG
        b'\x89PNG',           # PNG
        b'GIF87a',            # GIF
        b'GIF89a',            # GIF
        b'BM',                # BMP
    ]
    is_image = False
    for header in image_headers:
        if raw_content.startswith(header):
            is_image = True
            break

    # 如果是图片，尝试提取嵌入的base64数据
    if is_image:
        import base64
        print(f"  检测到图片文件，尝试提取嵌入的base64数据...")
        try:
            text_content = raw_content.decode('latin-1')
            # 查找长base64字符串（至少50个字符）
            base64_pattern = r'[A-Za-z0-9+/=]{50,}'
            match = re.search(base64_pattern, text_content)
            if match:
                base64_str = match.group(0)
                print(f"  找到base64数据，长度: {len(base64_str)}")
                decoded = base64.b64decode(base64_str)
                content = decoded.decode('utf-8')
                # 移除JavaScript风格的注释
                content = re.sub(r'^//.*$', '', content, flags=re.MULTILINE).strip()
                return content
        except Exception as e:
            print(f"  提取图片数据失败: {e}")

    # 普通文本文件处理
    content = raw_content.decode('utf-8')
    # 移除JavaScript风格的注释
    content = re.sub(r'^//.*$', '', content, flags=re.MULTILINE).strip()
    return content

# 下载 spider 文件
def extract_and_save_spider(json_text):
    match = re.search(r'"spider"\s*:\s*"([^"]+)"', json_text)
    if not match:
        raise ValueError("未找到 spider 字段")
    full_spider = match.group(1)
    spider_url = full_spider.split(";")[0]
    print(f"📥 下载 spider 文件: {spider_url}")
    resp = session.get(spider_url, timeout=30, allow_redirects=True)
    with open("fan.txt", "wb") as f:
        f.write(resp.content)
    print("✅ 已保存为 fan.txt")

# 计算本地文件 MD5
def get_md5(filepath):
    md5 = hashlib.md5()
    with open(filepath, "rb") as f:
        while chunk := f.read(8192):
            md5.update(chunk)
    return md5.hexdigest()

# 删除不需要的 sites 项 + 替换链接
def clean_data(raw_text):
    # 统一把各种 GitHub 代理壳替换掉
    raw_text = re.sub(
        r'https?://[^/]+/https://raw\.githubusercontent\.com/fantaiying7/EXT/refs/heads/main',
        './FTY',
        raw_text
    )

    data = demjson.decode(raw_text)

    # keywords = [
    #     "豆", "饭太硬", "广告", "PanSso", "YpanSo", "xzso", "米搜", "夸搜", "Aliso", "YiSo"
    # ]

    # original_count = len(data.get("sites", []))

    # data["sites"] = [
    #     s for s in data["sites"]
    #     if not any(kw in s.get("key", "") or kw in s.get("name", "") for kw in keywords)
    # ]

    # print(f"🧹 清理 {data - len(data['sites'])} 条 sites")
    return data

# 格式美化保存
class CompactJSONEncoder(json.JSONEncoder):
    def iterencode(self, o, _one_shot=False):
        def _compact_list(lst, indent_level):
            pad = '  ' * indent_level
            if all(isinstance(i, dict) for i in lst):
                return '[\n' + ',\n'.join([pad + '  ' + json.dumps(i, ensure_ascii=False, separators=(',', ': ')) for i in lst]) + '\n' + pad + ']'
            return json.dumps(lst, ensure_ascii=False, indent=2)
        def _encode(obj, indent_level=0):
            pad = '  ' * indent_level
            if isinstance(obj, dict):
                lines = [f'"{k}": {_encode(v, indent_level+1)}' for k, v in obj.items()]
                return '{\n' + pad + '  ' + (',\n' + pad + '  ').join(lines) + '\n' + pad + '}'
            elif isinstance(obj, list):
                return _compact_list(obj, indent_level)
            return json.dumps(obj, ensure_ascii=False)
        return iter([_encode(o)])

def save_json(data, filename="tvbox_cleaned.json"):
    with open(filename, "w", encoding="utf-8") as f:
        json.dump(data, f, ensure_ascii=False, indent=2, cls=CompactJSONEncoder)
    print(f"✅ 最终 JSON 保存为 {filename}")

# 主流程
if __name__ == "__main__":
    try:
        # 判断输入源
        if len(sys.argv) > 1:
            input_path = sys.argv[1]
            if input_path in ['-h', '--help']:
                print("用法:")
                print("  python fty.py [输入文件]")
                print("  python fty.py              # 从URL获取数据")
                print("  python fty.py input.json   # 从本地JSON文件读取")
                print("  python fty.py fff.json     # 从本地图片文件读取（自动提取base64数据）")
                sys.exit(0)
            if os.path.isfile(input_path):
                print(f"📂 读取本地文件: {input_path}")
                raw_text = read_local_json(input_path)
            else:
                print(f"❌ 文件不存在: {input_path}")
                sys.exit(1)
        else:
            print("🌐 从URL获取数据...")
            raw_text = fetch_raw_json()

        extract_and_save_spider(raw_text)
        data = clean_data(raw_text)
        # 更新 spider 为本地 fan.txt + 最新 MD5
        md5_value = get_md5("fan.txt")
        data["spider"] = f"./jar/fan.txt;md5;{md5_value}"
        print(f"🔄 spider 已更新为: {data['spider']}")
        save_json(data)
    except Exception as e:
        print(f"❌ 错误: {e}")