v0.2.7: 修复Redis连接 + 启动管理后台

- 修复Redis认证 (配置密码)
- 启动Python管理后台 (端口9531, 15个功能开关)
- 统一版本号 0.2.7
- 更新docker-compose.yml (镜像版本/Redis URL/Admin服务)

cloudsearch_transfer/adapter/baidu/__init__.py

@@ -0,0 +1,253 @@
+"""
+百度网盘适配器 — CloudSearch Transfer v1.0.0
+参考 cloud-auto-save 的 BaiduNetDisk + netdisk 的 PanbaiduSave
+
+完整的 5 步转存流程 + bdstoken 管理 + 路径删除 + 广告过滤
+"""
+
+import logging
+from typing import List, Tuple
+
+from ..base import BaseCloudDriveAdapter, FileInfo
+from ...config import PlatformConfig, TransferConfig
+from ...errors import TransferError, TransferErrorCode
+
+from .credential import BaiduCredentialManager
+from .transfer import BaiduTransfer
+from .cleanup import BaiduCleanup
+
+logger = logging.getLogger(__name__)
+
+
+class BaiduAdapter(BaseCloudDriveAdapter):
+    """百度网盘适配器
+
+    完整的 Cookie + bdstoken 机制，支持:
+    - 验证分享链接 + 提取码
+    - 5 步转存到自己的网盘
+    - 创建新分享
+    - 按文件名删除文件
+    - 广告文件过滤
+    """
+
+    PLATFORM_NAME = "百度网盘"
+    PLATFORM_KEY = "baidu"
+    URL_PATTERNS = [
+        r'pan\.baidu\.com/s/1([A-Za-z0-9_-]+)',
+    ]
+
+    def __init__(self, config: PlatformConfig, transfer_config: TransferConfig):
+        super().__init__(config, transfer_config)
+
+        # 凭证管理器
+        self.credential = BaiduCredentialManager(
+            cookie=config.cookie,
+            session=self.session,
+        )
+
+        if not self.credential.validate():
+            raise TransferError(
+                TransferErrorCode.NOT_LOGIN,
+                message="百度网盘 Cookie 无效或太短 (需 >= 50 字符)",
+                platform=self.PLATFORM_KEY,
+            )
+
+        # 预热 bdstoken
+        try:
+            self.credential.get_bdstoken()
+        except TransferError as e:
+            logger.warning(f"预取 bdstoken 失败: {e}，将在首次使用时重试")
+
+        # 转存执行器 & 清理器
+        self._transfer = BaiduTransfer(self.session, self.credential)
+        self._cleanup = BaiduCleanup(
+            self.session, self.credential,
+            ad_keywords=config.banned_keywords or None,
+        )
+
+        # 暂存最近一次转存的文件信息（供 _filter_ads 使用）
+        self._last_transfer_files: List[dict] = []
+
+    # ─── session 初始化 ─────────────────────────────────────
+
+    def _setup_session(self):
+        """设置 session 级别的 Cookie"""
+        if self.config.cookie:
+            self.session.headers["Cookie"] = self.config.cookie
+        self.session.headers["Referer"] = "https://pan.baidu.com/"
+
+    # ─── 核心抽象方法实现 ──────────────────────────────────
+
+    def _get_share_detail(self, pwd_id: str, passcode: str = "") -> dict:
+        """获取百度分享详情（步骤 ①+②）
+
+        Args:
+            pwd_id:   URL 中的 surl (s/1 后面的部分)
+            passcode: 提取码（可选）
+
+        Returns:
+            {"title": str, "fs_ids": [str], "filenames": [str], ...}
+        """
+        bdstoken = self.credential.get_bdstoken()
+
+        # ① 验证提取码（如果有）
+        if passcode:
+            self._transfer._verify_password(pwd_id, passcode, bdstoken)
+
+        # ② 解析分享页
+        share_info = self._transfer._parse_share_page(pwd_id)
+
+        return {
+            "title": share_info.get("title", ""),
+            "shareid": share_info["shareid"],
+            "uk": share_info["uk"],
+            "fs_ids": share_info["fs_ids"],
+            "filenames": share_info["filenames"],
+        }
+
+    def _save_files(self, pwd_id: str, detail: dict,
+                    save_dir: str) -> List[str]:
+        """转存文件到自己的百度网盘（步骤 ③+④）
+
+        Args:
+            pwd_id:   surl
+            detail:   _get_share_detail 返回的 dict
+            save_dir: 目标目录
+
+        Returns:
+            转存后的新 fs_id 列表
+        """
+        bdstoken = self.credential.get_bdstoken()
+        shareid = detail["shareid"]
+        uk = detail["uk"]
+        fs_ids = detail["fs_ids"]
+        filenames = detail.get("filenames", [])
+
+        # ③ 转存
+        self._transfer._transfer_files(shareid, uk, fs_ids, save_dir, bdstoken)
+
+        # ④ 列出目录匹配新 fs_id
+        new_fs_ids = self._transfer._list_and_match(save_dir, filenames, bdstoken)
+
+        # 暂存文件信息供 _filter_ads + _create_share 使用
+        self._last_transfer_files = [
+            {"fs_id": fid, "name": name}
+            for fid, name in zip(new_fs_ids, filenames)
+            if fid
+        ]
+
+        return new_fs_ids
+
+    def _create_share(self, file_ids: List[str], title: str,
+                      password: str = "") -> Tuple[str, str]:
+        """创建百度分享（步骤 ⑤）
+
+        Args:
+            file_ids: 转存后的新 fs_id 列表
+            title:    原标题
+            password: 分享密码
+
+        Returns:
+            (new_share_url, share_password)
+        """
+        # 如果 file_ids 中包含非数字，尝试从暂存信息中查找
+        numeric_ids = []
+        for fid in file_ids:
+            try:
+                int(fid)
+                numeric_ids.append(fid)
+            except ValueError:
+                logger.warning(f"忽略非数字 fs_id: {fid}")
+
+        return self._transfer.create_share(
+            fids=[int(x) for x in numeric_ids] if numeric_ids else [int(x) for x in file_ids],
+            password=password,
+            period=0,  # 永久
+        )
+
+    # ─── 文件列表 & 删除 ────────────────────────────────────
+
+    def get_files(self, parent_fid: str = "0") -> List[FileInfo]:
+        """列出百度网盘目录下的文件
+
+        GET /api/list?dir={parent_fid}
+
+        Args:
+            parent_fid: 目录路径 (默认 "0" = 根目录)
+
+        注意: parent_fid 对百度网盘而言是目录路径而非数字 ID。
+              根目录传 "/" 或 "0"。
+        """
+        bdstoken = self.credential.get_bdstoken()
+        dir_path = parent_fid if parent_fid != "0" else "/"
+
+        url = "https://pan.baidu.com/api/list"
+        params = {"dir": dir_path, "bdstoken": bdstoken}
+        headers = self.credential.get_headers()
+
+        try:
+            resp = self._get(url, params=params, headers=headers)
+            data = resp.json()
+        except Exception as e:
+            logger.error(f"百度列出目录失败: {e}")
+            return []
+
+        errno = data.get("errno", -1)
+        if errno != 0:
+            logger.error(f"百度列出目录 errno={errno}: {data}")
+            return []
+
+        files = []
+        for item in data.get("list", []):
+            fid = str(item.get("fs_id", ""))
+            name = item.get("server_filename", "")
+            size = item.get("size", 0)
+            is_dir = item.get("isdir", 0) == 1
+            ext = ""
+            if not is_dir and "." in name:
+                ext = name.rsplit(".", 1)[-1]
+
+            files.append(FileInfo(
+                fid=fid,
+                name=name,
+                size=size,
+                is_dir=is_dir,
+                ext=ext,
+            ))
+
+        return files
+
+    def delete(self, file_ids: List[str]) -> bool:
+        """删除百度网盘文件（按路径）
+
+        file_ids 应为网盘中的完整路径，如 ["/dir/file.txt", "/dir/file2.zip"]
+
+        Args:
+            file_ids: 网盘路径列表
+
+        Returns:
+            True 删除成功（或文件不存在）
+        """
+        return self._cleanup.delete_files(file_ids)
+
+    # ─── 广告过滤 ────────────────────────────────────────────
+
+    def _filter_ads(self, file_ids: List[str]) -> List[str]:
+        """广告过滤 — 基于最近一次转存暂存的文件名"""
+        if not self._last_transfer_files:
+            return file_ids
+
+        names = []
+        for f in self._last_transfer_files:
+            if f["fs_id"] in file_ids:
+                names.append(f["name"])
+            else:
+                names.append("")
+
+        return self._cleanup.filter_ad_ids(file_ids, names)
+
+    # ─── 扩展方法 ────────────────────────────────────────────
+
+    def delete_paths(self, paths: List[str]) -> bool:
+        """便捷删除方法（直接调用 cleanup）"""
+        return self._cleanup.delete_files(paths)

cloudsearch_transfer/adapter/baidu/cleanup.py

@@ -0,0 +1,154 @@
+"""
+百度网盘文件清理 — 删除文件 & 广告过滤
+参考 cloud-auto-save 的 filter_ads + netdisk 的 delete
+"""
+
+import json
+import logging
+from typing import List
+
+import requests
+
+from ...errors import TransferError, TransferErrorCode
+from .credential import BaiduCredentialManager, BAIDU_PAN_API
+
+logger = logging.getLogger(__name__)
+
+# 默认广告关键词
+DEFAULT_AD_KEYWORDS = [
+    "公众号", "微信", "扫码", "加群", "QQ群", "广告",
+    "关注", "免费领取", "点击领取", "全网", "最全",
+    "防走丢", "防迷路", "备用", "务必下载", "必看",
+    "解压密码", "压缩密码",
+]
+
+
+class BaiduCleanup:
+    """百度网盘文件清理 & 广告过滤"""
+
+    def __init__(self, session: requests.Session,
+                 credential: BaiduCredentialManager,
+                 ad_keywords: List[str] = None):
+        self.session = session
+        self.credential = credential
+        self.ad_keywords = ad_keywords or DEFAULT_AD_KEYWORDS
+
+    # ─── 删除文件 ────────────────────────────────────────────
+
+    def delete_files(self, paths: List[str]) -> bool:
+        """批量删除文件（按网盘路径）
+
+        POST /api/filemanager?opera=delete&bdstoken={bdstoken}
+        Body: filelist=["/path/to/file1","/path/to/file2"]
+
+        Args:
+            paths: 文件在网盘中的完整路径列表，如 ["/dir/file.txt"]
+
+        Returns:
+            True 全部成功（包括文件不存在的 errno=2）
+
+        Raises:
+            TransferError: 删除失败
+        """
+        if not paths:
+            logger.info("删除列表为空，跳过")
+            return True
+
+        bdstoken = self.credential.get_bdstoken()
+        url = f"{BAIDU_PAN_API}/api/filemanager"
+        params = {
+            "opera": "delete",
+            "bdstoken": bdstoken,
+        }
+        data = {
+            "filelist": json.dumps(paths, ensure_ascii=False),
+        }
+        headers = self.credential.get_headers()
+        headers["Content-Type"] = "application/x-www-form-urlencoded"
+
+        try:
+            resp = self.session.post(
+                url, params=params, data=data, headers=headers, timeout=30
+            )
+            resp.raise_for_status()
+            result = resp.json()
+        except Exception as e:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"百度删除请求失败: {e}",
+                platform="baidu",
+            )
+
+        errno = result.get("errno", -1)
+
+        # errno=0 成功; errno=2 文件不存在（视为成功）
+        if errno in (0, 2):
+            logger.info(f"百度删除完成: {len(paths)} 个路径 (errno={errno})")
+            return True
+
+        raise TransferError(
+            TransferErrorCode.NETWORK_ERROR,
+            message=f"百度删除失败 (errno={errno})",
+            platform="baidu",
+            details=result,
+        )
+
+    # ─── 广告过滤 ────────────────────────────────────────────
+
+    def filter_ads(self, files: List[dict]) -> List[dict]:
+        """根据文件名过滤广告文件
+
+        Args:
+            files: [{"fs_id": "xxx", "name": "xxx"}, ...]
+
+        Returns:
+            过滤后的文件列表，仅保留非广告文件
+        """
+        if not self.ad_keywords:
+            return files
+
+        retained = []
+        removed = []
+        for f in files:
+            name = f.get("name", "")
+            if self._is_ad(name):
+                removed.append(name)
+            else:
+                retained.append(f)
+
+        if removed:
+            logger.info(f"广告过滤: 移除 {len(removed)} 个文件: {removed}")
+        return retained
+
+    def filter_ad_ids(self, file_ids: List[str],
+                      file_names: List[str]) -> List[str]:
+        """根据文件名过滤广告，返回保留的 file_ids
+
+        Args:
+            file_ids:  文件 ID 列表
+            file_names: 对应的文件名列表（与 file_ids 一一对应）
+
+        Returns:
+            过滤后的 file_ids
+        """
+        if not self.ad_keywords:
+            return file_ids
+
+        retained = []
+        for fid, name in zip(file_ids, file_names):
+            if not self._is_ad(name):
+                retained.append(fid)
+            else:
+                logger.info(f"广告过滤: 移除 {name}")
+
+        return retained
+
+    def _is_ad(self, filename: str) -> bool:
+        """判断文件名是否为广告"""
+        if not filename:
+            return False
+        name_lower = filename.lower()
+        for kw in self.ad_keywords:
+            if kw.lower() in name_lower:
+                return True
+        return False

cloudsearch_transfer/adapter/baidu/credential.py

@@ -0,0 +1,101 @@
+"""
+百度网盘凭证管理器 — bdstoken 获取与校验
+参考 cloud-auto-save 的 BaiduNetDisk.cookie 机制
+"""
+
+import logging
+import requests
+
+from ...errors import TransferError, TransferErrorCode
+
+logger = logging.getLogger(__name__)
+
+# 百度网盘 API 基础 URL
+BAIDU_PAN_API = "https://pan.baidu.com"
+
+
+class BaiduCredentialManager:
+    """百度网盘 Cookie 凭证 + bdstoken 管理
+
+    百度网盘的大多数受保护 API 都需要 bdstoken 参数，
+    该 token 通过 API 获取并缓存在实例中。
+    """
+
+    def __init__(self, cookie: str, session: requests.Session):
+        """
+        Args:
+            cookie: 完整的百度 Cookie 字符串
+            session: 共享的 requests.Session（继承 User-Agent 等 headers）
+        """
+        self.cookie = cookie
+        self.session = session
+        self._bdstoken: str = ""
+
+    # ─── 公开方法 ──────────────────────────────────────────
+
+    def validate(self) -> bool:
+        """校验 Cookie 是否有效：长度 >= 50 视为合格"""
+        return bool(self.cookie and len(self.cookie.strip()) >= 50)
+
+    def get_bdstoken(self, force_refresh: bool = False) -> str:
+        """
+        获取 bdstoken，首次调用会请求 API 获取并缓存。
+
+        API: GET /api/gettemplatevariable?fields=["bdstoken"]
+
+        Raises:
+            TransferError: 获取失败 (BAIDU_BDSTOKEN_FAIL)
+        """
+        if self._bdstoken and not force_refresh:
+            return self._bdstoken
+
+        url = f"{BAIDU_PAN_API}/api/gettemplatevariable"
+        params = {"fields": '["bdstoken"]'}
+        headers = self.get_headers()
+
+        try:
+            resp = self.session.get(url, params=params, headers=headers, timeout=15)
+            resp.raise_for_status()
+            data = resp.json()
+        except Exception as e:
+            logger.error(f"获取 bdstoken 网络异常: {e}")
+            raise TransferError(
+                TransferErrorCode.BAIDU_BDSTOKEN_FAIL,
+                message=f"百度 bdstoken 请求失败: {e}",
+                platform="baidu",
+            )
+
+        errno = data.get("errno", -1)
+        if errno != 0:
+            logger.error(f"获取 bdstoken API 返回 errno={errno}: {data}")
+            raise TransferError(
+                TransferErrorCode.BAIDU_BDSTOKEN_FAIL,
+                message=f"百度 bdstoken 获取失败 (errno={errno})",
+                platform="baidu",
+                details={"response": data},
+            )
+
+        self._bdstoken = data.get("result", {}).get("bdstoken", "")
+        if not self._bdstoken:
+            raise TransferError(
+                TransferErrorCode.BAIDU_BDSTOKEN_FAIL,
+                message="百度 bdstoken 为空",
+                platform="baidu",
+            )
+
+        logger.info("bdstoken 获取成功")
+        return self._bdstoken
+
+    def get_headers(self) -> dict:
+        """构建携带 Cookie 的请求头（继承 session 默认 headers 外的额外字段）"""
+        headers = {
+            "Cookie": self.cookie,
+            "Referer": "https://pan.baidu.com/",
+            "Origin": "https://pan.baidu.com",
+        }
+        return headers
+
+    def invalidate_bdstoken(self):
+        """使缓存失效，下次调用 get_bdstoken 会重新获取"""
+        self._bdstoken = ""
+        logger.info("bdstoken 缓存已失效")

cloudsearch_transfer/adapter/baidu/transfer.py

@@ -0,0 +1,448 @@
+"""
+百度网盘转存核心 — 5 步转存流程
+参考 netdisk 的 PanbaiduSave + cloud-auto-save 的 BaiduNetDisk.transfer
+
+流程:
+  ① 验证提取码 → POST /share/verify
+  ② 解析分享页 → GET /s/1{surl}
+  ③ 转存文件 → POST /share/transfer
+  ④ 列出目录 → GET /api/list
+  ⑤ 创建分享 → POST /share/set
+"""
+
+import re
+import json
+import logging
+from typing import List, Tuple
+
+import requests
+
+from ...errors import TransferError, TransferErrorCode
+from .credential import BaiduCredentialManager, BAIDU_PAN_API
+
+logger = logging.getLogger(__name__)
+
+# ─── 正则 ──────────────────────────────────────────────────
+
+# 从 HTML 中提取 shareid
+RE_SHAREID = re.compile(r"""shareid["\s:=]+(\d+)""")
+# 从 HTML 中提取 uk
+RE_UK = re.compile(r"""uk["\s:=]+(\d+)""")
+# 从 HTML 中提取 fs_id
+RE_FS_ID = re.compile(r'"fs_id"\s*:\s*(\d+)')
+# 从 HTML 中提取 server_filename
+RE_FILENAME = re.compile(r'"server_filename"\s*:\s*"([^"]*)"')
+# 从 HTML/JSON 中提取标题
+RE_TITLE = re.compile(r'"title"\s*:\s*"([^"]*)"')
+# 从 HTML 中提取文件列表 JSON 块 (file_list 对象) — 标记位置
+RE_FILE_LIST_MARK = re.compile(r'"file_list"\s*:\s*(\{)', re.DOTALL)
+# 提取单个文件条目 (fallback)
+RE_FILE_ENTRY = re.compile(r'\{"fs_id":(\d+),"server_filename":"([^"]+)"')
+
+
+class BaiduTransfer:
+    """百度网盘 5 步转存执行器
+
+    每个实例绑定一个 Session + Cookie + bdstoken，
+    执行完整的「验证→解析→转存→查目录→创建分享」流程。
+    """
+
+    def __init__(self, session: requests.Session,
+                 credential: BaiduCredentialManager):
+        self.session = session
+        self.credential = credential
+        self.cookie = credential.cookie
+
+    # ─── 5 步主流程 ────────────────────────────────────────
+
+    def execute(self, surl: str, password: str,
+                save_dir: str = "/") -> Tuple[List[str], dict]:
+        """执行完整的 5 步转存流程
+
+        Args:
+            surl:   分享短码 (s/1 后面的部分)
+            password: 提取码
+            save_dir: 转存目标目录
+
+        Returns:
+            (new_fs_ids, file_info_dict)
+            new_fs_ids:  转存后的文件 fs_id 列表
+            file_info_dict: {fs_id: name} 映射
+
+        Raises:
+            TransferError: 任何一步失败
+        """
+        bdstoken = self.credential.get_bdstoken()
+
+        # ① 验证提取码
+        logger.info(f"[百度转存] ① 验证提取码 surl={surl}")
+        self._verify_password(surl, password, bdstoken)
+
+        # ② 解析分享页
+        logger.info(f"[百度转存] ② 解析分享页 surl={surl}")
+        share_info = self._parse_share_page(surl)
+        shareid = share_info["shareid"]
+        uk = share_info["uk"]
+        fs_ids = share_info["fs_ids"]
+        filenames = share_info["filenames"]
+        title = share_info.get("title", "")
+
+        if not fs_ids:
+            raise TransferError(
+                TransferErrorCode.RESOURCE_EMPTY,
+                message="分享中没有找到可转存的文件",
+                platform="baidu",
+            )
+
+        # ③ 转存到自己的网盘
+        logger.info(f"[百度转存] ③ 转存 {len(fs_ids)} 个文件到 {save_dir}")
+        self._transfer_files(shareid, uk, fs_ids, save_dir, bdstoken)
+
+        # ④ 列出目标目录，按文件名匹配新的 fs_id
+        logger.info(f"[百度转存] ④ 列出目录 {save_dir} 匹配新 fs_id")
+        new_fs_ids = self._list_and_match(save_dir, filenames, bdstoken)
+
+        if not new_fs_ids:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message="转存后无法匹配到新文件 ID",
+                platform="baidu",
+            )
+
+        # 构建返回的 info dict
+        file_info = {}
+        for name, fid in zip(filenames, new_fs_ids) if len(filenames) == len(new_fs_ids) else []:
+            file_info[fid] = name
+        if not file_info:
+            for fid in new_fs_ids:
+                file_info[fid] = title or fid
+
+        return new_fs_ids, file_info
+
+    def create_share(self, fids: List[int], password: str = "",
+                     period: int = 0) -> Tuple[str, str]:
+        """⑤ 创建新分享
+
+        Args:
+            fids:     转存后的文件 fs_id 列表
+            password: 分享密码（空 = 无密码）
+            period:   分享有效期 (0=永久)
+
+        Returns:
+            (share_url, share_password)
+        """
+        bdstoken = self.credential.get_bdstoken()
+        url = f"{BAIDU_PAN_API}/share/set"
+        params = {
+            "channel": "chunlei",
+            "clienttype": "0",
+            "web": "1",
+            "bdstoken": bdstoken,
+        }
+        data = {
+            "fid_list": json.dumps(fids),
+            "period": period,
+            "pwd": password,
+        }
+        headers = self.credential.get_headers()
+
+        try:
+            resp = self.session.post(
+                url, params=params, data=data, headers=headers, timeout=30
+            )
+            resp.raise_for_status()
+        except Exception as e:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"创建分享请求失败: {e}",
+                platform="baidu",
+            )
+
+        result = resp.json()
+        errno = result.get("errno", -1)
+
+        if errno == 9219:
+            raise TransferError(
+                TransferErrorCode.SHARE_LIMIT,
+                message="百度今日分享次数过多",
+                platform="baidu",
+            )
+        if errno != 0:
+            raise TransferError(
+                TransferErrorCode.SHARE_LINK_FAIL,
+                message=f"创建分享失败 (errno={errno})",
+                platform="baidu",
+                details=result,
+            )
+
+        share_url = result.get("link", "")
+        share_password = result.get("pwd", password) or password
+
+        logger.info(f"[百度转存] ⑤ 分享创建成功: {share_url}")
+        return share_url, share_password
+
+    # ─── 5 步内部方法 ──────────────────────────────────────
+
+    def _verify_password(self, surl: str, password: str, bdstoken: str):
+        """① 验证提取码
+
+        POST /share/verify?surl={surl}&bdstoken={bdstoken}
+        Body: {"pwd": "xxxx"}
+
+        errno=0 表示通过；errno=-9 表示提取码错误；errno=2 表示分享不存在
+        """
+        url = f"{BAIDU_PAN_API}/share/verify"
+        params = {
+            "surl": surl,
+            "bdstoken": bdstoken,
+        }
+        data = {"pwd": password}
+        headers = self.credential.get_headers()
+        headers["Content-Type"] = "application/x-www-form-urlencoded"
+
+        try:
+            resp = self.session.post(
+                url, params=params, data=data, headers=headers, timeout=15
+            )
+            resp.raise_for_status()
+        except Exception as e:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"验证提取码请求失败: {e}",
+                platform="baidu",
+            )
+
+        result = resp.json()
+        errno = result.get("errno", -1)
+
+        if errno == 0:
+            logger.info("提取码验证通过")
+            return
+
+        if errno == -9 or errno == -62:
+            raise TransferError(
+                TransferErrorCode.PASSCODE_WRONG,
+                message="百度提取码错误",
+                platform="baidu",
+            )
+        if errno == 2 or errno == 118:
+            raise TransferError(
+                TransferErrorCode.SHARE_NOT_EXIST,
+                message="百度分享不存在或已失效",
+                platform="baidu",
+            )
+        raise TransferError(
+            TransferErrorCode.NETWORK_ERROR,
+            message=f"验证提取码失败 (errno={errno})",
+            platform="baidu",
+            details=result,
+        )
+
+    def _parse_share_page(self, surl: str) -> dict:
+        """② 解析分享页面 HTML
+
+        GET /s/1{surl}
+        从 HTML 中正则提取 shareid, uk, fs_id[], server_filename[]
+        """
+        url = f"{BAIDU_PAN_API}/s/1{surl}"
+        headers = self.credential.get_headers()
+
+        try:
+            resp = self.session.get(url, headers=headers, timeout=20)
+            resp.raise_for_status()
+            html = resp.text
+        except Exception as e:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"打开分享页面失败: {e}",
+                platform="baidu",
+            )
+
+        # 提取 shareid
+        m_shareid = RE_SHAREID.search(html)
+        if not m_shareid:
+            raise TransferError(
+                TransferErrorCode.SHARE_NOT_EXIST,
+                message="无法从页面中提取 shareid，分享可能已失效",
+                platform="baidu",
+            )
+        shareid = m_shareid.group(1)
+
+        # 提取 uk
+        m_uk = RE_UK.search(html)
+        uk = m_uk.group(1) if m_uk else ""
+
+        # 提取标题
+        m_title = RE_TITLE.search(html)
+        title = m_title.group(1) if m_title else ""
+
+        # 提取文件列表 — 优先从 file_list JSON 块中提取
+        fs_ids = []
+        filenames = []
+
+        # 方法1：查找 file_list JSON 块（使用括号计数提取平衡 JSON）
+        m_fl = RE_FILE_LIST_MARK.search(html)
+        if m_fl:
+            start = m_fl.start(1)  # { 的位置
+            depth = 1
+            end = start + 1
+            while end < len(html) and depth > 0:
+                if html[end] == '{':
+                    depth += 1
+                elif html[end] == '}':
+                    depth -= 1
+                end += 1
+            file_list_json = html[start:end]
+            try:
+                file_list = json.loads(file_list_json)
+                for entry in file_list.get("list", []):
+                    fs_ids.append(str(entry.get("fs_id", "")))
+                    filenames.append(entry.get("server_filename", ""))
+            except json.JSONDecodeError:
+                pass
+
+        # 方法2：退化为正则提取所有 fs_id + server_filename
+        if not fs_ids:
+            for m in RE_FILE_ENTRY.finditer(html):
+                fs_ids.append(m.group(1))
+                filenames.append(m.group(2))
+
+        if not fs_ids:
+            # 可能只有一个文件，尝试单个提取
+            m_fsid = RE_FS_ID.search(html)
+            m_name = RE_FILENAME.search(html)
+            if m_fsid:
+                fs_ids.append(m_fsid.group(1))
+                filenames.append(m_name.group(1) if m_name else "")
+
+        logger.info(
+            f"解析分享页: shareid={shareid}, uk={uk}, "
+            f"文件数={len(fs_ids)}, title={title[:30]}"
+        )
+        return {
+            "shareid": shareid,
+            "uk": uk,
+            "fs_ids": fs_ids,
+            "filenames": filenames,
+            "title": title,
+        }
+
+    def _transfer_files(self, shareid: str, uk: str,
+                        fs_ids: List[str], save_dir: str, bdstoken: str):
+        """③ 转存文件到自己的网盘
+
+        POST /share/transfer?shareid={shareid}&from={uk}&bdstoken={bdstoken}
+        Body: fsidlist=[1,2,3]&path=/dir
+        """
+        url = f"{BAIDU_PAN_API}/share/transfer"
+        params = {
+            "shareid": shareid,
+            "from": uk,
+            "bdstoken": bdstoken,
+        }
+        data = {
+            "fsidlist": json.dumps([int(x) for x in fs_ids]),
+            "path": save_dir,
+        }
+        headers = self.credential.get_headers()
+        headers["Content-Type"] = "application/x-www-form-urlencoded"
+
+        try:
+            resp = self.session.post(
+                url, params=params, data=data, headers=headers, timeout=30
+            )
+            resp.raise_for_status()
+        except Exception as e:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"转存请求失败: {e}",
+                platform="baidu",
+            )
+
+        result = resp.json()
+        errno = result.get("errno", -1)
+
+        if errno == 0:
+            logger.info(f"转存成功: {len(fs_ids)} 个文件 → {save_dir}")
+            return
+
+        if errno == 12:
+            raise TransferError(
+                TransferErrorCode.CAPACITY_FULL,
+                message="百度网盘空间不足",
+                platform="baidu",
+            )
+        if errno == 9013:
+            raise TransferError(
+                TransferErrorCode.SENSITIVE_RESOURCE,
+                message="文件包含违规内容，无法转存",
+                platform="baidu",
+            )
+        raise TransferError(
+            TransferErrorCode.NETWORK_ERROR,
+            message=f"转存失败 (errno={errno})",
+            platform="baidu",
+            details=result,
+        )
+
+    def _list_and_match(self, save_dir: str, filenames: List[str],
+                        bdstoken: str) -> List[str]:
+        """④ 列出目标目录，按文件名匹配新的 fs_id
+
+        GET /api/list?dir={dir}&bdstoken={bdstoken}
+        从返回的 list 中按 server_filename 匹配，返回按原顺序排列的 fs_id 列表
+        """
+        url = f"{BAIDU_PAN_API}/api/list"
+        params = {
+            "dir": save_dir,
+            "bdstoken": bdstoken,
+        }
+        headers = self.credential.get_headers()
+
+        try:
+            resp = self.session.get(url, params=params, headers=headers, timeout=15)
+            resp.raise_for_status()
+            data = resp.json()
+        except Exception as e:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"列出目录失败: {e}",
+                platform="baidu",
+            )
+
+        errno = data.get("errno", -1)
+        if errno == -12:
+            raise TransferError(
+                TransferErrorCode.DIR_NOT_EXIST,
+                message=f"百度目录不存在: {save_dir}",
+                platform="baidu",
+            )
+        if errno != 0:
+            raise TransferError(
+                TransferErrorCode.NETWORK_ERROR,
+                message=f"列出目录失败 (errno={errno})",
+                platform="baidu",
+                details=data,
+            )
+
+        file_list = data.get("list", [])
+        # 构建文件名 → fs_id 映射
+        name_to_fid = {}
+        for item in file_list:
+            name = item.get("server_filename", "")
+            fid = str(item.get("fs_id", ""))
+            if name and fid:
+                name_to_fid[name] = fid
+
+        # 按原文件名顺序匹配
+        new_fs_ids = []
+        for fname in filenames:
+            if fname in name_to_fid:
+                new_fs_ids.append(name_to_fid[fname])
+            else:
+                logger.warning(f"目录中未找到文件: {fname}")
+
+        logger.info(
+            f"目录匹配: 期望 {len(filenames)} 个, 匹配到 {len(new_fs_ids)} 个"
+        )
+        return new_fs_ids