diff --git a/.gitignore b/.gitignore index 3531e17..da38698 100644 --- a/.gitignore +++ b/.gitignore @@ -1,3 +1,4 @@ downloads/ __pycache__/ *.pyc +.env diff --git a/README.md b/README.md index cf97972..0ac7f82 100644 --- a/README.md +++ b/README.md @@ -7,7 +7,7 @@ - ✅ **自动绕过反爬保护**：自动解决网站的 SHA1 工作量证明（PoW）挑战 - ✅ **智能搜索匹配**：支持书名、作者、ISBN 等多维度匹配 - ✅ **格式优先级**：自动选择 epub + pdf 版本下载 -- ✅ **代理轮询**：支持 Shadowsocks（Surge 格式）代理列表，每 IP 5 本后自动切换 +- ✅ **代理轮询**：支持 Shadowsocks（Surge 格式）代理列表，或直接通过 Surge `select` 组切换出口 - ✅ **AI 友好**：JSON 输出，结构化搜索结果，适合 AI Agent 调用 ## 安装 @@ -23,6 +23,8 @@ pip install curl_cffi beautifulsoup4 lxml 脚本内置自动代理引擎配置。如果使用 Shadowsocks 代理列表（如 Surge 格式），脚本**会在首次运行时自动将对应平台架构的 `mihomo` 引擎打包进执行环境中**，完全脱离 Host 依赖，无需手动通过 brew 安装任何插件。 +如果你已经在 macOS 上使用 Surge，并且将 zlib 域名路由到一个专门的 `select` 组，也可以直接使用 Surge HTTP API 轮换该组，无需再让脚本自行启动 `mihomo`。 + ## 使用方法 ### 搜索图书 @@ -55,6 +57,9 @@ python3 zlib_dl.py download --title "Project Hail Mary" --author "Andy Weir" -o # 使用代理列表 python3 zlib_dl.py download --title "Project Hail Mary" --author "Andy Weir" --proxy-file proxies.txt + +# 使用 Surge 的 Z-Library 策略组轮换出口（推荐先配置 .env） +python3 zlib_dl.py download --title "Project Hail Mary" --author "Andy Weir" ``` ### 详细模式 @@ -91,6 +96,25 @@ http://127.0.0.1:8080 - 达到上限后自动停止当前代理进程，启动下一个 - SS 代理会自动启动 `mihomo` 进程创建本地隧道，无需宿主机环境准备 +### Surge 策略组模式 + +- 为 Surge 打开 HTTP API，例如：`http-api = @127.0.0.1:6171` +- 将 zlib 相关域名规则指向一个 `select` 组，例如 `Z-Library` +- 脚本会默认通过 `http://127.0.0.1:6152` 访问 Surge 本地 HTTP 代理 +- 当命中 IP 限额或下载失败时，脚本会调用 Surge API，把 `Z-Library` 切到下一个策略 + +在项目根目录创建 `.env`： + +```env +SURGE_API_URL=http://127.0.0.1:6171 +SURGE_API_KEY=6f8d2c1a9b4e7f35c0d2a8e1b7f94c63 +SURGE_GROUP=Z-Library +SURGE_HTTP_PROXY=http://127.0.0.1:6152 +SURGE_SOCKS_PROXY=socks5://127.0.0.1:6153 +``` + +CLI 参数优先级高于 `.env`；如果不传 `--surge-api`，但 `.env` 里存在 `SURGE_API_URL`，下载命令会自动进入 Surge 模式。 + ## AI Agent 集成 ### JSON 搜索结果结构 diff --git a/zlib_dl.py b/zlib_dl.py index 16161c8..ef96b3f 100644 --- a/zlib_dl.py +++ b/zlib_dl.py @@ -24,6 +24,8 @@ import sys import time import unicodedata import urllib.request +import urllib.parse +import urllib.error from dataclasses import dataclass, field, asdict from pathlib import Path from typing import Optional, List, Dict, Any, Tuple @@ -139,6 +141,14 @@ class ProxyConfig: _process: Any = field(default=None, repr=False) +@dataclass +class SurgeConfig: + api_url: str + api_key: str + group_name: str + proxy_url: str + + # ============================================================ # PoW 挑战解决器 # ============================================================ @@ -178,6 +188,26 @@ class ChallengeSolver: return None +def load_dotenv(dotenv_path: str = ".env"): + """加载本地 .env 文件，不覆盖已存在的环境变量""" + if not os.path.exists(dotenv_path): + return + + try: + with open(dotenv_path, "r", encoding="utf-8") as f: + for raw_line in f: + line = raw_line.strip() + if not line or line.startswith("#") or "=" not in line: + continue + key, value = line.split("=", 1) + key = key.strip() + value = value.strip().strip('"').strip("'") + if key and key not in os.environ: + os.environ[key] = value + except Exception: + pass + + # ============================================================ # ZLib 会话管理 # ============================================================ @@ -222,14 +252,14 @@ class ZLibSession: ) ip = parser(resp.json()) print( - f" 🌐 [{stage}] 当前出口 IP: {ip} (proxy={proxy_desc}, probe={probe_url})", + f" 🌐 [{stage}] 调试探针出口 IP: {ip} (proxy={proxy_desc}, probe={probe_url})", file=sys.stderr, ) return except Exception as e: last_error = e - print(f" 🌐 [{stage}] 当前出口 IP 获取失败: {last_error}", file=sys.stderr) + print(f" 🌐 [{stage}] 调试探针出口 IP 获取失败: {last_error}", file=sys.stderr) def _request(self, url: str, **kwargs) -> cffi_requests.Response: """发起请求，自动处理 PoW 挑战""" @@ -485,6 +515,173 @@ class ZLibSession: # 代理管理器 # ============================================================ +class SurgeRuntime: + """通过 Surge HTTP API 轮换指定策略组""" + + MAX_DOWNLOADS_PER_PROXY = 5 + + def __init__(self, config: SurgeConfig, verbose: bool = False): + self.config = config + self.verbose = verbose + self.policy_names: List[str] = [] + self.current_index = -1 + self.downloads_count = 0 + + def _api_get(self, path: str, params: Optional[Dict[str, str]] = None) -> Any: + url = urllib.parse.urljoin(self.config.api_url.rstrip("/") + "/", path.lstrip("/")) + if params: + url = f"{url}?{urllib.parse.urlencode(params)}" + req = urllib.request.Request(url, headers={"X-Key": self.config.api_key, "Accept": "application/json"}) + with urllib.request.urlopen(req, timeout=10) as resp: + return json.loads(resp.read().decode("utf-8")) + + def _api_post(self, path: str, payload: Dict[str, Any]) -> Any: + url = urllib.parse.urljoin(self.config.api_url.rstrip("/") + "/", path.lstrip("/")) + data = json.dumps(payload).encode("utf-8") + req = urllib.request.Request( + url, + data=data, + method="POST", + headers={ + "X-Key": self.config.api_key, + "Accept": "application/json", + "Content-Type": "application/json", + }, + ) + with urllib.request.urlopen(req, timeout=10) as resp: + return json.loads(resp.read().decode("utf-8")) + + def _extract_group_policies(self, payload: Any) -> List[str]: + def _coerce_names(items: Any) -> List[str]: + names: List[str] = [] + if isinstance(items, list): + for item in items: + if isinstance(item, str): + names.append(item) + elif isinstance(item, dict) and item.get("name"): + names.append(str(item["name"])) + return names + + if isinstance(payload, dict): + direct = payload.get(self.config.group_name) + names = _coerce_names(direct) + if names: + return names + + groups = payload.get("policy_groups") + if isinstance(groups, list): + for group in groups: + if isinstance(group, dict) and group.get("name") == self.config.group_name: + names = _coerce_names(group.get("policies") or group.get("options") or []) + if names: + return names + + groups = payload.get("groups") + if isinstance(groups, dict): + group = groups.get(self.config.group_name) + if isinstance(group, dict): + names = _coerce_names(group.get("policies") or group.get("options") or []) + if names: + return names + names = _coerce_names(group) + if names: + return names + + raise RuntimeError(f"无法从 Surge API 响应中解析策略组 {self.config.group_name}") + + def _load_policies(self) -> List[str]: + data = self._api_get("/v1/policy_groups") + policies = self._extract_group_policies(data) + if not policies: + raise RuntimeError(f"策略组 {self.config.group_name} 中没有可用代理") + self.policy_names = policies + return policies + + def _sync_current_index(self): + current = self.get_current_proxy_name(refresh=True) + if current in self.policy_names: + self.current_index = self.policy_names.index(current) + + def has_proxy_pool(self) -> bool: + return True + + def ensure_active(self) -> bool: + try: + if not self.policy_names: + self._load_policies() + self._sync_current_index() + return self.current_index >= 0 + except Exception as e: + print(f"代理错误: 初始化 Surge 策略组失败: {e}", file=sys.stderr) + return False + + def get_current_proxy_url(self) -> Optional[str]: + return self.config.proxy_url + + def get_current_proxy_name(self, refresh: bool = True) -> str: + try: + data = self._api_get("/v1/policy_groups/select", {"group_name": self.config.group_name}) if refresh else None + if data is not None: + policy = data.get("policy") + if isinstance(policy, str) and policy: + if policy in self.policy_names: + self.current_index = self.policy_names.index(policy) + return policy + except Exception: + pass + + if 0 <= self.current_index < len(self.policy_names): + return self.policy_names[self.current_index] + return self.config.group_name + + def record_download(self): + self.downloads_count += 1 + + def should_rotate(self) -> bool: + return self.downloads_count >= self.MAX_DOWNLOADS_PER_PROXY + + def rotate(self) -> bool: + try: + if not self.policy_names: + self._load_policies() + self._sync_current_index() + + if not self.policy_names: + return False + + next_index = self.current_index + 1 + if next_index >= len(self.policy_names): + if self.verbose: + print(" ⚠️ Surge 策略组中的代理已用完", file=sys.stderr) + return False + + next_policy = self.policy_names[next_index] + self._api_post( + "/v1/policy_groups/select", + {"group_name": self.config.group_name, "policy": next_policy}, + ) + time.sleep(1.0) + self.current_index = next_index + self.downloads_count = 0 + + if self.verbose: + print(f" 📡 Surge 已切换策略组 {self.config.group_name} -> {next_policy}", file=sys.stderr) + return True + except Exception as e: + print(f"代理错误: 切换 Surge 策略组失败: {e}", file=sys.stderr) + return False + + def remaining_proxies(self) -> int: + if not self.policy_names: + return 0 + return max(0, len(self.policy_names) - self.current_index - 1) + + def remaining_downloads(self) -> int: + return max(0, self.MAX_DOWNLOADS_PER_PROXY - self.downloads_count) + + def cleanup(self): + return None + class ProxyManager: """管理代理列表和轮询""" @@ -1124,7 +1321,7 @@ def _should_rotate_proxy(message: str) -> bool: def _create_session_for_current_proxy( - proxy_mgr: ProxyManager, + proxy_mgr: Any, explicit_proxy: Optional[str], verbose: bool, ) -> ZLibSession: @@ -1137,20 +1334,47 @@ def cmd_download(args): """执行下载命令""" explicit_proxy = getattr(args, "proxy", None) proxy_file = getattr(args, "proxy_file", None) - if explicit_proxy and proxy_file: - print("错误: --proxy 与 --proxy-file 不能同时使用", file=sys.stderr) + surge_key = getattr(args, "surge_key", None) or os.environ.get("SURGE_API_KEY") + surge_api = getattr(args, "surge_api", None) or os.environ.get("SURGE_API_URL") + use_surge = bool(surge_api) + + if len([item for item in [explicit_proxy, proxy_file, surge_api] if item]) > 1: + print("错误: --proxy / --proxy-file / --surge-api 只能选择一种模式", file=sys.stderr) + sys.exit(1) + + if use_surge and not surge_key: + print("错误: Surge 模式需要 --surge-key 或环境变量 SURGE_API_KEY", file=sys.stderr) sys.exit(1) output_dir = Path(args.output_dir) output_dir.mkdir(parents=True, exist_ok=True) - # 代理管理器 - proxy_mgr = ProxyManager( - proxy_file=proxy_file, verbose=args.verbose - ) + if use_surge: + surge_key_str = str(surge_key) + surge_proxy_url = ( + getattr(args, "surge_http_proxy", None) + or os.environ.get("SURGE_HTTP_PROXY") + or getattr(args, "surge_socks_proxy", None) + or os.environ.get("SURGE_SOCKS_PROXY") + or "http://127.0.0.1:6152" + ) + proxy_mgr = SurgeRuntime( + config=SurgeConfig( + api_url=surge_api, + api_key=surge_key_str, + group_name=getattr(args, "surge_group", None) or os.environ.get("SURGE_GROUP", "Z-Library"), + proxy_url=surge_proxy_url, + ), + verbose=args.verbose, + ) + explicit_proxy = None + else: + proxy_mgr = ProxyManager( + proxy_file=proxy_file, verbose=args.verbose + ) if proxy_mgr.has_proxy_pool() and not proxy_mgr.ensure_active(): - print("错误: 代理池不可用，无法启动任何代理", file=sys.stderr) + print("错误: 代理不可用，无法初始化当前代理运行时", file=sys.stderr) sys.exit(1) download_results: List[DownloadResult] = [] @@ -1278,7 +1502,7 @@ def _download_one( author: str, fmt: str, output_dir: Path, - proxy_mgr: ProxyManager, + proxy_mgr: Any, explicit_proxy: Optional[str], verbose: bool, ) -> DownloadResult: @@ -1432,6 +1656,9 @@ def main(): # 使用代理下载 %(prog)s download --title "Project Hail Mary" --author "Andy Weir" --proxy-file proxies.txt + + # 使用 Surge 策略组轮换下载 + %(prog)s download --title "Project Hail Mary" --author "Andy Weir" --surge-api http://127.0.0.1:6171 """, ) @@ -1464,7 +1691,13 @@ def main(): dl_p.add_argument("--output-dir", "-o", default="./downloads", help="下载目录 (默认 ./downloads)") dl_p.add_argument("--proxy-file", "-p", default=None, help="代理列表文件") dl_p.add_argument("--proxy", default=None, help="单个代理 URL (socks5://host:port)") + dl_p.add_argument("--surge-api", default=None, help="Surge HTTP API 地址 (如 http://127.0.0.1:6171)") + dl_p.add_argument("--surge-key", default=None, help="Surge HTTP API Key，也可用环境变量 SURGE_API_KEY") + dl_p.add_argument("--surge-group", default=None, help="Surge select 组名 (默认读取 .env 或 Z-Library)") + dl_p.add_argument("--surge-http-proxy", default=None, help="Surge 本地 HTTP 代理 (默认读取 .env 或 http://127.0.0.1:6152)") + dl_p.add_argument("--surge-socks-proxy", default=None, help="Surge 本地 SOCKS5 代理 (如 socks5://127.0.0.1:6153)") + load_dotenv() args = parser.parse_args() if not args.command: