feat 支持豆瓣做为识别源

README.md

@@ -78,9 +78,11 @@ MoviePilot需要配套下载器和媒体服务器配合使用。
 - **TMDB_API_DOMAIN：** TMDB API地址，默认`api.themoviedb.org`，也可配置为`api.tmdb.org`或其它中转代理服务地址，能连通即可
 - **TMDB_IMAGE_DOMAIN：** TMDB图片地址，默认`image.tmdb.org`，可配置为其它中转代理以加速TMDB图片显示，如：`static-mdb.v.geilijiasu.com`
 - **WALLPAPER：** 登录首页电影海报，`tmdb`/`bing`，默认`tmdb`
+- **SEARCH_SOURCE：** 媒体信息搜索来源，`themoviedb`/`douban`，默认`themoviedb`
+- **RECOGNIZE_SOURCE：** 媒体信息识别来源，`themoviedb`/`douban`，默认`themoviedb`
+- **SCRAP_SOURCE：** 刮削元数据及图片使用的数据源，`themoviedb`/`douban`，默认`themoviedb`
 ---
 - **SCRAP_METADATA：** 刮削入库的媒体文件，`true`/`false`，默认`true`
-- **SCRAP_SOURCE：** 刮削元数据及图片使用的数据源，`themoviedb`/`douban`，默认`themoviedb`
 - **SCRAP_FOLLOW_TMDB：** 新增已入库媒体是否跟随TMDB信息变化，`true`/`false`，默认`true`
 ---
 - **❗TRANSFER_TYPE：** 整理转移方式，支持`link`/`copy`/`move`/`softlink`/`rclone_copy`/`rclone_move`  **注意：在`link`和`softlink`转移方式下，转移后的文件会继承源文件的权限掩码，不受`UMASK`影响；rclone需要自行映射rclone配置目录到容器中或在容器内完成rclone配置，节点名称必须为：`MP`**
@@ -100,7 +102,6 @@ MoviePilot需要配套下载器和媒体服务器配合使用。
 - **SUBSCRIBE_MODE：** 订阅模式，`rss`/`spider`，默认`spider`，`rss`模式通过定时刷新RSS来匹配订阅（RSS地址会自动获取，也可手动维护），对站点压力小，同时可设置订阅刷新周期，24小时运行，但订阅和下载通知不能过滤和显示免费，推荐使用rss模式。
 - **SUBSCRIBE_RSS_INTERVAL：** RSS订阅模式刷新时间间隔（分钟），默认`30`分钟，不能小于5分钟。
 - **SUBSCRIBE_SEARCH：** 订阅搜索，`true`/`false`，默认`false`，开启后会每隔24小时对所有订阅进行全量搜索，以补齐缺失剧集（一般情况下正常订阅即可，订阅搜索只做为兜底，会增加站点压力，不建议开启）。
-- **SEARCH_SOURCE：** 媒体信息搜索来源，`themoviedb`/`douban`，默认`themoviedb`
 - **AUTO_DOWNLOAD_USER：** 远程交互搜索时自动择优下载的用户ID，多个用户使用,分割，未设置需要选择资源或者回复`0`
 ---
 - **OCR_HOST：** OCR识别服务器地址，格式：`http(s)://ip:port`，用于识别站点验证码实现自动登录获取Cookie等，不配置默认使用内建服务器`https://movie-pilot.org`，可使用 [这个镜像](https://hub.docker.com/r/jxxghp/moviepilot-ocr) 自行搭建。

app/api/endpoints/douban.py

@@ -28,20 +28,6 @@ def douban_img(imgurl: str) -> Any:
     return None
 
 
-@router.get("/recognize/{doubanid}", summary="豆瓣ID识别", response_model=schemas.Context)
-def recognize_doubanid(doubanid: str,
-                       _: schemas.TokenPayload = Depends(verify_token)) -> Any:
-    """
-    根据豆瓣ID识别媒体信息
-    """
-    # 识别媒体信息
-    context = DoubanChain().recognize_by_doubanid(doubanid=doubanid)
-    if context:
-        return context.to_dict()
-    else:
-        return schemas.Context()
-
-
 @router.get("/showing", summary="豆瓣正在热映", response_model=List[schemas.MediaInfo])
 def movie_showing(page: int = 1,
                   count: int = 30,

app/api/endpoints/download.py

@@ -3,14 +3,13 @@ from typing import Any, List
 from fastapi import APIRouter, Depends, HTTPException
 
 from app import schemas
-from app.db.models.user import User
-from app.db.userauth import get_current_active_user
-from app.chain.douban import DoubanChain
 from app.chain.download import DownloadChain
 from app.chain.media import MediaChain
 from app.core.context import MediaInfo, Context, TorrentInfo
 from app.core.metainfo import MetaInfo
 from app.core.security import verify_token
+from app.db.models.user import User
+from app.db.userauth import get_current_active_user
 from app.schemas import NotExistMediaInfo, MediaType
 
 router = APIRouter()
@@ -61,20 +60,13 @@ def exists(media_in: schemas.MediaInfo,
     查询缺失媒体信息
     """
     # 媒体信息
-    mediainfo = MediaInfo()
     meta = MetaInfo(title=media_in.title)
-    if media_in.tmdb_id:
-        mediainfo.from_dict(media_in.dict())
-    elif media_in.douban_id:
-        context = DoubanChain().recognize_by_doubanid(doubanid=media_in.douban_id)
-        if context:
-            mediainfo = context.media_info
-            meta = context.meta_info
+    mtype = MediaType(media_in.type) if media_in.type else None
+    if media_in.tmdb_id or media_in.douban_id:
+        mediainfo = MediaChain().recognize_media(meta=meta, mtype=mtype,
+                                                 tmdbid=media_in.tmdb_id, doubanid=media_in.douban_id)
     else:
-        context = MediaChain().recognize_by_title(title=f"{media_in.title} {media_in.year}")
-        if context:
-            mediainfo = context.media_info
-            meta = context.meta_info
+        mediainfo = MediaChain().recognize_by_title(title=f"{media_in.title} {media_in.year}")
     # 查询缺失信息
     if not mediainfo or not mediainfo.tmdb_id:
         raise HTTPException(status_code=404, detail="媒体信息不存在")

app/api/endpoints/media.py

@@ -4,10 +4,8 @@ from fastapi import APIRouter, Depends
 from sqlalchemy.orm import Session
 
 from app import schemas
-from app.chain.douban import DoubanChain
 from app.chain.media import MediaChain
-from app.chain.tmdb import TmdbChain
-from app.core.context import MediaInfo
+from app.core.config import settings
 from app.core.metainfo import MetaInfo
 from app.core.security import verify_token
 from app.db import get_db
@@ -81,26 +79,30 @@ def exists(title: str = None,
 
 
 @router.get("/{mediaid}", summary="查询媒体详情", response_model=schemas.MediaInfo)
-def tmdb_info(mediaid: str, type_name: str,
-              _: schemas.TokenPayload = Depends(verify_token)) -> Any:
+def media_info(mediaid: str, type_name: str,
+               _: schemas.TokenPayload = Depends(verify_token)) -> Any:
     """
     根据媒体ID查询themoviedb或豆瓣媒体信息，type_name: 电影/电视剧
     """
     mtype = MediaType(type_name)
+    tmdbid, doubanid = None, None
     if mediaid.startswith("tmdb:"):
-        result = TmdbChain().tmdb_info(int(mediaid[5:]), mtype)
-        return MediaInfo(tmdb_info=result).to_dict()
+        tmdbid = int(mediaid[5:])
     elif mediaid.startswith("douban:"):
-        # 查询豆瓣信息
-        doubaninfo = DoubanChain().douban_info(doubanid=mediaid[7:])
-        if not doubaninfo:
-            return schemas.MediaInfo()
-        result = DoubanChain().recognize_by_doubaninfo(doubaninfo)
-        if result:
-            # TMDB
-            return result.media_info.to_dict()
-        else:
-            # 豆瓣
-            return MediaInfo(douban_info=doubaninfo).to_dict()
-    else:
+        doubanid = mediaid[7:]
+    if not tmdbid and not doubanid:
         return schemas.MediaInfo()
+    if settings.RECOGNIZE_SOURCE == "themoviedb":
+        if not tmdbid and doubanid:
+            tmdbinfo = MediaChain().get_tmdbinfo_by_doubanid(doubanid=doubanid, mtype=mtype)
+            if tmdbinfo:
+                tmdbid = tmdbinfo.get("id")
+    else:
+        if not doubanid and tmdbid:
+            doubaninfo = MediaChain().get_doubaninfo_by_tmdbid(tmdbid=tmdbid, mtype=mtype)
+            if doubaninfo:
+                doubanid = doubaninfo.get("id")
+    mediainfo = MediaChain().recognize_media(tmdbid=tmdbid, doubanid=doubanid, mtype=mtype)
+    if mediainfo:
+        return mediainfo.to_dict()
+    return schemas.MediaInfo()

app/api/endpoints/search.py

@@ -3,8 +3,9 @@ from typing import List, Any
 from fastapi import APIRouter, Depends
 
 from app import schemas
-from app.chain.douban import DoubanChain
+from app.chain.media import MediaChain
 from app.chain.search import SearchChain
+from app.core.config import settings
 from app.core.security import verify_token
 from app.schemas.types import MediaType
 
@@ -21,27 +22,36 @@ async def search_latest(_: schemas.TokenPayload = Depends(verify_token)) -> Any:
 
 
 @router.get("/media/{mediaid}", summary="精确搜索资源", response_model=List[schemas.Context])
-def search_by_tmdbid(mediaid: str,
-                     mtype: str = None,
-                     area: str = "title",
-                     _: schemas.TokenPayload = Depends(verify_token)) -> Any:
+def search_by_id(mediaid: str,
+                 mtype: str = None,
+                 area: str = "title",
+                 _: schemas.TokenPayload = Depends(verify_token)) -> Any:
     """
     根据TMDBID/豆瓣ID精确搜索站点资源 tmdb:/douban:/
     """
+    torrents = []
+    if mtype:
+        mtype = MediaType(mtype)
     if mediaid.startswith("tmdb:"):
         tmdbid = int(mediaid.replace("tmdb:", ""))
-        if mtype:
-            mtype = MediaType(mtype)
-        torrents = SearchChain().search_by_tmdbid(tmdbid=tmdbid, mtype=mtype, area=area)
+        if settings.RECOGNIZE_SOURCE == "douban":
+            # 通过TMDBID识别豆瓣ID
+            doubaninfo = MediaChain().get_doubaninfo_by_tmdbid(tmdbid=tmdbid, mtype=mtype)
+            if doubaninfo:
+                torrents = SearchChain().search_by_id(doubanid=doubaninfo.get("id"),
+                                                      mtype=mtype, area=area)
+        else:
+            torrents = SearchChain().search_by_id(tmdbid=tmdbid, mtype=mtype, area=area)
     elif mediaid.startswith("douban:"):
         doubanid = mediaid.replace("douban:", "")
-        # 识别豆瓣信息
-        context = DoubanChain().recognize_by_doubanid(doubanid)
-        if not context or not context.media_info or not context.media_info.tmdb_id:
-            return []
-        torrents = SearchChain().search_by_tmdbid(tmdbid=context.media_info.tmdb_id,
-                                                  mtype=context.media_info.type,
-                                                  area=area)
+        if settings.RECOGNIZE_SOURCE == "themoviedb":
+            # 通过豆瓣ID识别TMDBID
+            tmdbinfo = MediaChain().get_tmdbinfo_by_doubanid(doubanid=doubanid, mtype=mtype)
+            if tmdbinfo:
+                torrents = SearchChain().search_by_id(tmdbid=tmdbinfo.get("id"),
+                                                      mtype=mtype, area=area)
+        else:
+            torrents = SearchChain().search_by_id(doubanid=doubanid, mtype=mtype, area=area)
     else:
         return []
     return [torrent.to_dict() for torrent in torrents]

app/chain/__init__.py

@@ -107,28 +107,34 @@ class ChainBase(metaclass=ABCMeta):
                     # 中止继续执行
                     break
             except Exception as err:
-                logger.error(f"运行模块 {method} 出错：{module.__class__.__name__} - {str(err)}\n{traceback.print_exc()}")
+                logger.error(
+                    f"运行模块 {method} 出错：{module.__class__.__name__} - {str(err)}\n{traceback.print_exc()}")
         return result
 
     def recognize_media(self, meta: MetaBase = None,
                         mtype: MediaType = None,
-                        tmdbid: int = None) -> Optional[MediaInfo]:
+                        tmdbid: int = None,
+                        doubanid: str = None) -> Optional[MediaInfo]:
         """
         识别媒体信息
         :param meta:     识别的元数据
         :param mtype:    识别的媒体类型，与tmdbid配套
         :param tmdbid:   tmdbid
+        :param doubanid: 豆瓣ID
         :return: 识别的媒体信息，包括剧集信息
         """
+        # 识别用名中含指定信息情形
+        if not mtype and meta and meta.type in [MediaType.TV, MediaType.MOVIE]:
+            mtype = meta.type
         if not tmdbid and hasattr(meta, "tmdbid"):
-            # 识别用名中含指定信息情形
             tmdbid = meta.tmdbid
-            if not mtype and meta.type in [MediaType.TV, MediaType.MOVIE]:
-                mtype = meta.type
-        return self.run_module("recognize_media", meta=meta, mtype=mtype, tmdbid=tmdbid)
+        if not doubanid and hasattr(meta, "doubanid"):
+            doubanid = meta.doubanid
+        return self.run_module("recognize_media", meta=meta, mtype=mtype,
+                               tmdbid=tmdbid, doubanid=doubanid)
 
     def match_doubaninfo(self, name: str, imdbid: str = None,
-                         mtype: str = None, year: str = None, season: int = None) -> Optional[dict]:
+                         mtype: MediaType = None, year: str = None, season: int = None) -> Optional[dict]:
         """
         搜索和匹配豆瓣信息
         :param name: 标题
@@ -140,6 +146,18 @@ class ChainBase(metaclass=ABCMeta):
         return self.run_module("match_doubaninfo", name=name, imdbid=imdbid,
                                mtype=mtype, year=year, season=season)
 
+    def match_tmdbinfo(self, name: str, mtype: MediaType = None,
+                       year: str = None, season: int = None) -> Optional[dict]:
+        """
+        搜索和匹配TMDB信息
+        :param name: 标题
+        :param mtype: 类型
+        :param year: 年份
+        :param season: 季
+        """
+        return self.run_module("match_tmdbinfo", name=name,
+                               mtype=mtype, year=year, season=season)
+
     def obtain_images(self, mediainfo: MediaInfo) -> Optional[MediaInfo]:
         """
         补充抓取媒体信息图片
@@ -164,13 +182,14 @@ class ChainBase(metaclass=ABCMeta):
                                image_prefix=image_prefix, image_type=image_type,
                                season=season, episode=episode)
 
-    def douban_info(self, doubanid: str) -> Optional[dict]:
+    def douban_info(self, doubanid: str, mtype: MediaType = None) -> Optional[dict]:
         """
         获取豆瓣信息
         :param doubanid: 豆瓣ID
+        :param mtype: 媒体类型
         :return: 豆瓣信息
         """
-        return self.run_module("douban_info", doubanid=doubanid)
+        return self.run_module("douban_info", doubanid=doubanid, mtype=mtype)
 
     def tvdb_info(self, tvdbid: int) -> Optional[dict]:
         """

app/chain/douban.py

@@ -1,10 +1,6 @@
 from typing import Optional, List
 
 from app.chain import ChainBase
-from app.core.context import Context
-from app.core.context import MediaInfo
-from app.core.metainfo import MetaInfo
-from app.log import logger
 from app.schemas import MediaType
 from app.utils.singleton import Singleton
 
@@ -14,52 +10,6 @@ class DoubanChain(ChainBase, metaclass=Singleton):
     豆瓣处理链，单例运行
     """
 
-    def recognize_by_doubanid(self, doubanid: str) -> Optional[Context]:
-        """
-        根据豆瓣ID识别媒体信息
-        """
-        logger.info(f'开始识别媒体信息，豆瓣ID：{doubanid} ...')
-        # 查询豆瓣信息
-        doubaninfo = self.douban_info(doubanid=doubanid)
-        if not doubaninfo:
-            logger.warn(f'未查询到豆瓣信息，豆瓣ID：{doubanid}')
-            return None
-        return self.recognize_by_doubaninfo(doubaninfo)
-
-    def recognize_by_doubaninfo(self, doubaninfo: dict) -> Optional[Context]:
-        """
-        根据豆瓣信息识别媒体信息
-        """
-        # 优先使用原标题匹配
-        season_meta = None
-        if doubaninfo.get("original_title"):
-            meta = MetaInfo(title=doubaninfo.get("original_title"))
-            season_meta = MetaInfo(title=doubaninfo.get("title"))
-            # 合并季
-            meta.begin_season = season_meta.begin_season
-        else:
-            meta = MetaInfo(title=doubaninfo.get("title"))
-        # 年份
-        if doubaninfo.get("year"):
-            meta.year = doubaninfo.get("year")
-        # 处理类型
-        if isinstance(doubaninfo.get('media_type'), MediaType):
-            meta.type = doubaninfo.get('media_type')
-        else:
-            meta.type = MediaType.MOVIE if doubaninfo.get("type") == "movie" else MediaType.TV
-        # 使用原标题识别媒体信息
-        mediainfo = self.recognize_media(meta=meta, mtype=meta.type)
-        if not mediainfo:
-            if season_meta and season_meta.name != meta.name:
-                # 使用主标题识别媒体信息
-                mediainfo = self.recognize_media(meta=season_meta, mtype=season_meta.type)
-            if not mediainfo:
-                logger.warn(f'{meta.name} 未识别到TMDB媒体信息')
-                return Context(meta_info=meta, media_info=MediaInfo(douban_info=doubaninfo))
-        logger.info(f'识别到媒体信息：{mediainfo.type.value} {mediainfo.title_year} {meta.season}')
-        mediainfo.set_douban_info(doubaninfo)
-        return Context(meta_info=meta, media_info=mediainfo)
-
     def movie_top250(self, page: int = 1, count: int = 30) -> List[dict]:
         """
         获取豆瓣电影TOP250

app/chain/download.py

@@ -682,6 +682,7 @@ class DownloadChain(ChainBase):
         if mediainfo.type == MediaType.MOVIE:
             # 电影
             itemid = self.mediaserver.get_item_id(mtype=mediainfo.type.value,
+                                                  title=mediainfo.title,
                                                   tmdbid=mediainfo.tmdb_id)
             exists_movies: Optional[ExistMediaInfo] = self.media_exists(mediainfo=mediainfo, itemid=itemid)
             if exists_movies:
@@ -692,7 +693,8 @@ class DownloadChain(ChainBase):
             if not mediainfo.seasons:
                 # 补充媒体信息
                 mediainfo: MediaInfo = self.recognize_media(mtype=mediainfo.type,
-                                                            tmdbid=mediainfo.tmdb_id)
+                                                            tmdbid=mediainfo.tmdb_id,
+                                                            doubanid=mediainfo.douban_id)
                 if not mediainfo:
                     logger.error(f"媒体信息识别失败！")
                     return False, {}
@@ -701,6 +703,7 @@ class DownloadChain(ChainBase):
                     return False, {}
             # 电视剧
             itemid = self.mediaserver.get_item_id(mtype=mediainfo.type.value,
+                                                  title=mediainfo.title,
                                                   tmdbid=mediainfo.tmdb_id,
                                                   season=mediainfo.season)
             # 媒体库已存在的剧集

app/chain/media.py

@@ -14,7 +14,6 @@ from app.schemas.types import EventType, MediaType
 from app.utils.singleton import Singleton
 from app.utils.string import StringUtils
 
-
 recognize_lock = Lock()
 
 
@@ -27,7 +26,7 @@ class MediaChain(ChainBase, metaclass=Singleton):
     # 临时识别结果 {title, name, year, season, episode}
     recognize_temp: Optional[dict] = None
 
-    def recognize_by_title(self, title: str, subtitle: str = None) -> Optional[Context]:
+    def recognize_by_title(self, title: str, subtitle: str = None) -> Optional[MediaInfo]:
         """
         根据主副标题识别媒体信息
         """
@@ -43,13 +42,13 @@ class MediaChain(ChainBase, metaclass=Singleton):
                 mediainfo = self.recognize_help(title=title, org_meta=metainfo)
             if not mediainfo:
                 logger.warn(f'{title} 未识别到媒体信息')
-                return Context(meta_info=metainfo)
+                return None
         # 识别成功
         logger.info(f'{title} 识别到媒体信息：{mediainfo.type.value} {mediainfo.title_year}')
         # 更新媒体图片
         self.obtain_images(mediainfo=mediainfo)
         # 返回上下文
-        return Context(meta_info=metainfo, media_info=mediainfo)
+        return mediainfo
 
     def recognize_help(self, title: str, org_meta: MetaBase) -> Optional[MediaInfo]:
         """
@@ -190,3 +189,78 @@ class MediaChain(ChainBase, metaclass=Singleton):
         logger.info(f"{content} 搜索到 {len(medias)} 条相关媒体信息")
         # 识别的元数据，媒体信息列表
         return meta, medias
+
+    def get_tmdbinfo_by_doubanid(self, doubanid: str, mtype: MediaType = None) -> Optional[dict]:
+        """
+        根据豆瓣ID获取TMDB信息
+        """
+        tmdbinfo = None
+        doubaninfo = self.douban_info(doubanid=doubanid, mtype=mtype)
+        if doubaninfo:
+            # 优先使用原标题匹配
+            season_meta = None
+            if doubaninfo.get("original_title"):
+                meta = MetaInfo(title=doubaninfo.get("original_title"))
+                season_meta = MetaInfo(title=doubaninfo.get("title"))
+                # 合并季
+                meta.begin_season = season_meta.begin_season
+            else:
+                meta = MetaInfo(title=doubaninfo.get("title"))
+            # 年份
+            if doubaninfo.get("year"):
+                meta.year = doubaninfo.get("year")
+            # 处理类型
+            if isinstance(doubaninfo.get('media_type'), MediaType):
+                meta.type = doubaninfo.get('media_type')
+            else:
+                meta.type = MediaType.MOVIE if doubaninfo.get("type") == "movie" else MediaType.TV
+            # 使用原标题识别TMDB媒体信息
+            tmdbinfo = self.match_tmdbinfo(
+                name=meta.name,
+                year=meta.year,
+                mtype=mtype or meta.type,
+                season=meta.begin_season
+            )
+            if not tmdbinfo:
+                if season_meta and season_meta.name != meta.name:
+                    # 使用主标题识别媒体信息
+                    tmdbinfo = self.match_tmdbinfo(
+                        name=season_meta.name,
+                        year=meta.year,
+                        mtype=mtype or meta.type,
+                        season=meta.begin_season
+                    )
+        return tmdbinfo
+
+    def get_doubaninfo_by_tmdbid(self, tmdbid: int,
+                                 mtype: MediaType = None, season: int = None) -> Optional[dict]:
+        """
+        根据TMDBID获取豆瓣信息
+        """
+        tmdbinfo = self.tmdb_info(tmdbid=tmdbid, mtype=mtype)
+        if tmdbinfo:
+            # 名称
+            name = tmdbinfo.get("title") or tmdbinfo.get("name")
+            # 年份
+            year = None
+            if tmdbinfo.get('release_date'):
+                year = tmdbinfo['release_date'][:4]
+            elif tmdbinfo.get('seasons') and season:
+                for seainfo in tmdbinfo['seasons']:
+                    # 季
+                    season_number = seainfo.get("season_number")
+                    if not season_number:
+                        continue
+                    air_date = seainfo.get("air_date")
+                    if air_date and season_number == season:
+                        year = air_date[:4]
+                        break
+            # IMDBID
+            imdbid = tmdbinfo.get("external_ids", {}).get("imdb_id")
+            return self.match_doubaninfo(
+                name=name,
+                year=year,
+                mtype=mtype,
+                imdbid=imdbid
+            )
+        return None

app/chain/search.py

@@ -31,14 +31,16 @@ class SearchChain(ChainBase):
         self.systemconfig = SystemConfigOper()
         self.torrenthelper = TorrentHelper()
 
-    def search_by_tmdbid(self, tmdbid: int, mtype: MediaType = None, area: str = "title") -> List[Context]:
+    def search_by_id(self, tmdbid: int = None, doubanid: str = None,
+                     mtype: MediaType = None, area: str = "title") -> List[Context]:
         """
-        根据TMDB ID搜索资源，精确匹配，但不不过滤本地存在的资源
+        根据TMDBID/豆瓣ID搜索资源，精确匹配，但不不过滤本地存在的资源
         :param tmdbid: TMDB ID
+        :param doubanid: 豆瓣 ID
         :param mtype: 媒体，电影 or 电视剧
         :param area: 搜索范围，title or imdbid
         """
-        mediainfo = self.recognize_media(tmdbid=tmdbid, mtype=mtype)
+        mediainfo = self.recognize_media(tmdbid=tmdbid, doubanid=doubanid, mtype=mtype)
         if not mediainfo:
             logger.error(f'{tmdbid} 媒体信息识别失败！')
             return []
@@ -96,7 +98,8 @@ class SearchChain(ChainBase):
         # 补充媒体信息
         if not mediainfo.names:
             mediainfo: MediaInfo = self.recognize_media(mtype=mediainfo.type,
-                                                        tmdbid=mediainfo.tmdb_id)
+                                                        tmdbid=mediainfo.tmdb_id,
+                                                        doubanid=mediainfo.douban_id)
             if not mediainfo:
                 logger.error(f'媒体信息识别失败！')
                 return []
@@ -154,6 +157,7 @@ class SearchChain(ChainBase):
         if mediainfo:
             self.progress.start(ProgressKey.Search)
             logger.info(f'开始匹配，总 {_total} 个资源 ...')
+            logger.info(f"标题：{mediainfo.title}，原标题：{mediainfo.original_title}，别名：{mediainfo.names}")
             self.progress.update(value=0, text=f'开始匹配，总 {_total} 个资源 ...', key=ProgressKey.Search)
             for torrent in torrents:
                 _count += 1

app/chain/subscribe.py

@@ -6,10 +6,11 @@ from datetime import datetime
 from typing import Dict, List, Optional, Union, Tuple
 
 from app.chain import ChainBase
-from app.chain.douban import DoubanChain
 from app.chain.download import DownloadChain
+from app.chain.media import MediaChain
 from app.chain.search import SearchChain
 from app.chain.torrents import TorrentsChain
+from app.core.config import settings
 from app.core.context import TorrentInfo, Context, MediaInfo
 from app.core.meta import MetaBase
 from app.core.metainfo import MetaInfo
@@ -51,29 +52,32 @@ class SubscribeChain(ChainBase):
         识别媒体信息并添加订阅
         """
         logger.info(f'开始添加订阅，标题：{title} ...')
-        metainfo = None
         mediainfo = None
-        if not tmdbid and doubanid:
-            # 将豆瓣信息转换为TMDB信息
-            context = DoubanChain().recognize_by_doubanid(doubanid)
-            if context:
-                metainfo = context.meta_info
-                mediainfo = context.media_info
+        metainfo = MetaInfo(title)
+        if year:
+            metainfo.year = year
+        if mtype:
+            metainfo.type = mtype
+        if season:
+            metainfo.type = MediaType.TV
+            metainfo.begin_season = season
+        # 识别媒体信息
+        if settings.RECOGNIZE_SOURCE == "themoviedb":
+            # TMDB识别模式
+            if not tmdbid and doubanid:
+                # 将豆瓣信息转换为TMDB信息
+                tmdbinfo = MediaChain().get_tmdbinfo_by_doubanid(doubanid=doubanid, mtype=mtype)
+                if tmdbinfo:
+                    mediainfo = MediaInfo(tmdb_info=tmdbinfo)
+            else:
+                # 识别TMDB信息
+                mediainfo = self.recognize_media(meta=metainfo, mtype=mtype, tmdbid=tmdbid)
         else:
-            # 识别元数据
-            metainfo = MetaInfo(title)
-            if year:
-                metainfo.year = year
-            if mtype:
-                metainfo.type = mtype
-            if season:
-                metainfo.type = MediaType.TV
-                metainfo.begin_season = season
-            # 识别媒体信息
-            mediainfo = self.recognize_media(meta=metainfo, mtype=mtype, tmdbid=tmdbid)
+            # 豆瓣识别模式
+            mediainfo = self.recognize_media(meta=metainfo, mtype=mtype, doubanid=doubanid)
         # 识别失败
-        if not mediainfo or not metainfo or not mediainfo.tmdb_id:
-            logger.warn(f'未识别到媒体信息，标题：{title}，tmdbid：{tmdbid}')
+        if not mediainfo:
+            logger.warn(f'未识别到媒体信息，标题：{title}，tmdbid：{tmdbid}，doubanid：{doubanid}')
             return None, "未识别到媒体信息"
         # 更新媒体图片
         self.obtain_images(mediainfo=mediainfo)
@@ -86,16 +90,17 @@ class SubscribeChain(ChainBase):
                 if not mediainfo.seasons:
                     # 补充媒体信息
                     mediainfo = self.recognize_media(mtype=mediainfo.type,
-                                                     tmdbid=mediainfo.tmdb_id)
+                                                     tmdbid=mediainfo.tmdb_id,
+                                                     doubanid=mediainfo.douban_id)
                     if not mediainfo:
                         logger.error(f"媒体信息识别失败！")
                         return None, "媒体信息识别失败"
                     if not mediainfo.seasons:
-                        logger.error(f"媒体信息中没有季集信息，标题：{title}，tmdbid：{tmdbid}")
+                        logger.error(f"媒体信息中没有季集信息，标题：{title}，tmdbid：{tmdbid}，doubanid：{doubanid}")
                         return None, "媒体信息中没有季集信息"
                 total_episode = len(mediainfo.seasons.get(season) or [])
                 if not total_episode:
-                    logger.error(f'未获取到总集数，标题：{title}，tmdbid：{tmdbid}')
+                    logger.error(f'未获取到总集数，标题：{title}，tmdbid：{tmdbid}, doubanid：{doubanid}')
                     return None, f"未获取到第 {season} 季的总集数"
                 kwargs.update({
                     'total_episode': total_episode
@@ -106,8 +111,7 @@ class SubscribeChain(ChainBase):
                     'lack_episode': kwargs.get('total_episode')
                 })
         # 添加订阅
-        sid, err_msg = self.subscribeoper.add(mediainfo, doubanid=doubanid,
-                                              season=season, username=username, **kwargs)
+        sid, err_msg = self.subscribeoper.add(mediainfo, season=season, username=username, **kwargs)
         if not sid:
             logger.error(f'{mediainfo.title_year} {err_msg}')
             if not exist_ok and message:
@@ -179,7 +183,9 @@ class SubscribeChain(ChainBase):
             meta.begin_season = subscribe.season or None
             meta.type = MediaType(subscribe.type)
             # 识别媒体信息
-            mediainfo: MediaInfo = self.recognize_media(meta=meta, mtype=meta.type, tmdbid=subscribe.tmdbid)
+            mediainfo: MediaInfo = self.recognize_media(meta=meta, mtype=meta.type,
+                                                        tmdbid=subscribe.tmdbid,
+                                                        doubanid=subscribe.doubanid)
             if not mediainfo:
                 logger.warn(f'未识别到媒体信息，标题：{subscribe.name}，tmdbid：{subscribe.tmdbid}')
                 continue
@@ -475,7 +481,8 @@ class SubscribeChain(ChainBase):
             meta.begin_season = subscribe.season or None
             meta.type = MediaType(subscribe.type)
             # 识别媒体信息
-            mediainfo: MediaInfo = self.recognize_media(meta=meta, mtype=meta.type, tmdbid=subscribe.tmdbid)
+            mediainfo: MediaInfo = self.recognize_media(meta=meta, mtype=meta.type,
+                                                        tmdbid=subscribe.tmdbid, doubanid=subscribe.doubanid)
             if not mediainfo:
                 logger.warn(f'未识别到媒体信息，标题：{subscribe.name}，tmdbid：{subscribe.tmdbid}')
                 continue
@@ -661,7 +668,8 @@ class SubscribeChain(ChainBase):
             meta.begin_season = subscribe.season or None
             meta.type = MediaType(subscribe.type)
             # 识别媒体信息
-            mediainfo: MediaInfo = self.recognize_media(meta=meta, mtype=meta.type, tmdbid=subscribe.tmdbid)
+            mediainfo: MediaInfo = self.recognize_media(meta=meta, mtype=meta.type,
+                                                        tmdbid=subscribe.tmdbid, doubanid=subscribe.doubanid)
             if not mediainfo:
                 logger.warn(f'未识别到媒体信息，标题：{subscribe.name}，tmdbid：{subscribe.tmdbid}')
                 continue

app/chain/tmdb.py

@@ -25,17 +25,21 @@ class TmdbChain(ChainBase, metaclass=Singleton):
         :param page:  页码
         :return: 媒体信息列表
         """
+        if settings.RECOGNIZE_SOURCE != "themoviedb":
+            return None
         return self.run_module("tmdb_discover", mtype=mtype,
                                sort_by=sort_by, with_genres=with_genres,
                                with_original_language=with_original_language,
                                page=page)
 
-    def tmdb_trending(self, page: int = 1) -> List[dict]:
+    def tmdb_trending(self, page: int = 1) -> Optional[List[dict]]:
         """
         TMDB流行趋势
         :param page: 第几页
         :return: TMDB信息列表
         """
+        if settings.RECOGNIZE_SOURCE != "themoviedb":
+            return None
         return self.run_module("tmdb_trending", page=page)
 
     def tmdb_seasons(self, tmdbid: int) -> List[schemas.TmdbSeason]:

app/chain/transfer.py

@@ -66,7 +66,8 @@ class TransferChain(ChainBase):
                     mtype = MediaType(downloadhis.type)
                     # 按TMDBID识别
                     mediainfo = self.recognize_media(mtype=mtype,
-                                                     tmdbid=downloadhis.tmdbid)
+                                                     tmdbid=downloadhis.tmdbid,
+                                                     doubanid=downloadhis.doubanid)
                 else:
                     # 非MoviePilot下载的任务，按文件识别
                     mediainfo = None
@@ -449,7 +450,7 @@ class TransferChain(ChainBase):
 
         def args_error():
             self.post_message(Notification(channel=channel,
-                                           title="请输入正确的命令格式：/redo [id] [tmdbid]|[类型]，"
+                                           title="请输入正确的命令格式：/redo [id] [tmdbid/豆瓣id]|[类型]，"
                                                  "[id]历史记录编号", userid=userid))
 
         if not arg_str:
@@ -464,31 +465,32 @@ class TransferChain(ChainBase):
         if not logid.isdigit():
             args_error()
             return
-        # TMDB ID
-        tmdb_strs = arg_strs[1].split('|')
-        tmdbid = tmdb_strs[0]
+        # TMDBID/豆瓣ID
+        id_strs = arg_strs[1].split('|')
+        media_id = id_strs[0]
         if not logid.isdigit():
             args_error()
             return
         # 类型
-        type_str = tmdb_strs[1] if len(tmdb_strs) > 1 else None
+        type_str = id_strs[1] if len(id_strs) > 1 else None
         if not type_str or type_str not in [MediaType.MOVIE.value, MediaType.TV.value]:
             args_error()
             return
         state, errmsg = self.re_transfer(logid=int(logid),
-                                         mtype=MediaType(type_str), tmdbid=int(tmdbid))
+                                         mtype=MediaType(type_str),
+                                         mediaid=media_id)
         if not state:
             self.post_message(Notification(channel=channel, title="手动整理失败",
                                            text=errmsg, userid=userid))
             return
 
-    def re_transfer(self, logid: int,
-                    mtype: MediaType = None, tmdbid: int = None) -> Tuple[bool, str]:
+    def re_transfer(self, logid: int, mtype: MediaType = None,
+                    mediaid: str = None) -> Tuple[bool, str]:
         """
         根据历史记录，重新识别转移，只支持简单条件
         :param logid: 历史记录ID
         :param mtype: 媒体类型
-        :param tmdbid: TMDB ID
+        :param mediaid: TMDB ID/豆瓣ID
         """
         # 查询历史记录
         history: TransferHistory = self.transferhis.get(logid)
@@ -501,13 +503,14 @@ class TransferChain(ChainBase):
             return False, f"源目录不存在：{src_path}"
         dest_path = Path(history.dest) if history.dest else None
         # 查询媒体信息
-        if mtype and tmdbid:
-            mediainfo = self.recognize_media(mtype=mtype, tmdbid=tmdbid)
+        if mtype and mediaid:
+            mediainfo = self.recognize_media(mtype=mtype, tmdbid=int(mediaid) if str(mediaid).isdigit() else None,
+                                             doubanid=mediaid)
         else:
             meta = MetaInfoPath(src_path)
             mediainfo = self.recognize_media(meta=meta)
         if not mediainfo:
-            return False, f"未识别到媒体信息，类型：{mtype.value}，tmdbid：{tmdbid}"
+            return False, f"未识别到媒体信息，类型：{mtype.value}，id：{mediaid}"
         # 重新执行转移
         logger.info(f"{src_path.name} 识别为：{mediainfo.title_year}")
         # 更新媒体图片

app/core/config.py

@@ -43,14 +43,16 @@ class Settings(BaseSettings):
     WALLPAPER: str = "tmdb"
     # 网络代理 IP:PORT
     PROXY_HOST: str = None
-    # 媒体信息搜索来源
+    # 媒体识别来源 themoviedb/douban
+    RECOGNIZE_SOURCE: str = "themoviedb"
+    # 媒体信息搜索来源 themoviedb/douban
     SEARCH_SOURCE: str = "themoviedb"
+    # 刮削来源 themoviedb/douban
+    SCRAP_SOURCE: str = "themoviedb"
     # 刮削入库的媒体文件
     SCRAP_METADATA: bool = True
     # 新增已入库媒体是否跟随TMDB信息变化
     SCRAP_FOLLOW_TMDB: bool = True
-    # 刮削来源
-    SCRAP_SOURCE: str = "themoviedb"
     # TMDB图片地址
     TMDB_IMAGE_DOMAIN: str = "image.tmdb.org"
     # TMDB API地址

app/core/context.py

@@ -414,7 +414,6 @@ class MediaInfo:
         # 豆瓣ID
         self.douban_id = str(info.get("id"))
         # 类型
-
         if not self.type:
             if isinstance(info.get('media_type'), MediaType):
                 self.type = info.get('media_type')
@@ -422,10 +421,10 @@ class MediaInfo:
                 self.type = MediaType.MOVIE if info.get("type") == "movie" else MediaType.TV
         # 标题
         if not self.title:
-            self.title = info.get("title")
             # 识别标题中的季
-            meta = MetaInfo(self.title)
+            meta = MetaInfo(info.get("title"))
             self.season = meta.begin_season
+            self.title = meta.name
         # 原语种标题
         if not self.original_title:
             self.original_title = info.get("original_title")
@@ -472,14 +471,22 @@ class MediaInfo:
             self.actors = info.get("actors") or []
         # 别名
         if not self.names:
-            self.names = info.get("aka") or []
+            akas = info.get("aka")
+            if akas:
+                self.names = [re.sub(r'\([港台豆友译名]+\)', "", aka) for aka in akas]
         # 剧集
         if self.type == MediaType.TV and not self.seasons:
             meta = MetaInfo(info.get("title"))
-            if meta.begin_season:
-                episodes_count = info.get("episodes_count")
-                if episodes_count:
-                    self.seasons[meta.begin_season] = list(range(1, episodes_count + 1))
+            season = meta.begin_season or 1
+            episodes_count = info.get("episodes_count")
+            if episodes_count:
+                self.seasons[season] = list(range(1, episodes_count + 1))
+        # 季年份
+        if self.type == MediaType.TV and not self.season_years:
+            season = self.season or 1
+            self.season_years = {
+                season: self.year
+            }
         # 剩余属性赋值
         for key, value in info.items():
             if not hasattr(self, key):

app/core/meta/metabase.py

@@ -59,6 +59,9 @@ class MetaBase(object):
     audio_encode: Optional[str] = None
     # 应用的识别词信息
     apply_words: Optional[List[str]] = None
+    # 附加信息
+    tmdbid: int = None
+    doubanid: str = None
 
     # 副标题解析
     _subtitle_flag = False

app/core/metainfo.py

@@ -36,6 +36,8 @@ def MetaInfo(title: str, subtitle: str = None) -> MetaBase:
     # 修正媒体信息
     if metainfo.get('tmdbid'):
         meta.tmdbid = metainfo['tmdbid']
+    if metainfo.get('doubanid'):
+        meta.tmdbid = metainfo['doubanid']
     if metainfo.get('type'):
         meta.type = metainfo['type']
     if metainfo.get('begin_season'):
@@ -93,6 +95,7 @@ def find_metainfo(title: str) -> Tuple[str, dict]:
     """
     metainfo = {
         'tmdbid': None,
+        'doubanid': None,
         'type': None,
         'begin_season': None,
         'end_season': None,
@@ -108,10 +111,14 @@ def find_metainfo(title: str) -> Tuple[str, dict]:
     if not results:
         return title, metainfo
     for result in results:
-        tmdbid = re.findall(r'(?<=tmdbid=)\d+', result)
         # 查找tmdbid信息
+        tmdbid = re.findall(r'(?<=tmdbid=)\d+', result)
         if tmdbid and tmdbid[0].isdigit():
             metainfo['tmdbid'] = tmdbid[0]
+        # 查找豆瓣id信息
+        doubanid = re.findall(r'(?<=doubanid=)\d+', result)
+        if doubanid and doubanid[0].isdigit():
+            metainfo['doubanid'] = doubanid[0]
         # 查找媒体类型
         mtype = re.findall(r'(?<=type=)\d+', result)
         if mtype:

app/db/mediaserver_oper.py

@@ -39,10 +39,12 @@ class MediaServerOper(DbOper):
             # 优先按TMDBID查
             item = MediaServerItem.exist_by_tmdbid(self._db, tmdbid=kwargs.get("tmdbid"),
                                                    mtype=kwargs.get("mtype"))
-        else:
+        elif kwargs.get("title"):
             # 按标题、类型、年份查
             item = MediaServerItem.exists_by_title(self._db, title=kwargs.get("title"),
                                                    mtype=kwargs.get("mtype"), year=kwargs.get("year"))
+        else:
+            return None
         if not item:
             return None
 

app/db/subscribe_oper.py

@@ -23,6 +23,7 @@ class SubscribeOper(DbOper):
                                   tmdbid=mediainfo.tmdb_id,
                                   imdbid=mediainfo.imdb_id,
                                   tvdbid=mediainfo.tvdb_id,
+                                  doubanid=mediainfo.douban_id,
                                   poster=mediainfo.get_poster_image(),
                                   backdrop=mediainfo.get_backdrop_image(),
                                   vote=mediainfo.vote_average,

app/modules/douban/__init__.py

@@ -9,6 +9,7 @@ from app.core.metainfo import MetaInfo
 from app.log import logger
 from app.modules import _ModuleBase
 from app.modules.douban.apiv2 import DoubanApi
+from app.modules.douban.douban_cache import DoubanCache
 from app.modules.douban.scraper import DoubanScraper
 from app.schemas.types import MediaType
 from app.utils.common import retry
@@ -18,10 +19,12 @@ from app.utils.system import SystemUtils
 class DoubanModule(_ModuleBase):
     doubanapi: DoubanApi = None
     scraper: DoubanScraper = None
+    cache: DoubanCache = None
 
     def init_module(self) -> None:
         self.doubanapi = DoubanApi()
         self.scraper = DoubanScraper()
+        self.cache = DoubanCache()
 
     def stop(self):
         pass
@@ -29,10 +32,87 @@ class DoubanModule(_ModuleBase):
     def init_setting(self) -> Tuple[str, Union[str, bool]]:
         pass
 
-    def douban_info(self, doubanid: str) -> Optional[dict]:
+    def recognize_media(self, meta: MetaBase = None,
+                        mtype: MediaType = None,
+                        doubanid: str = None,
+                        **kwargs) -> Optional[MediaInfo]:
+        """
+        识别媒体信息
+        :param meta:     识别的元数据
+        :param mtype:    识别的媒体类型，与doubanid配套
+        :param doubanid: 豆瓣ID
+        :return: 识别的媒体信息，包括剧集信息
+        """
+        if settings.RECOGNIZE_SOURCE != "douban":
+            return None
+
+        if not meta:
+            cache_info = {}
+        else:
+            if mtype:
+                meta.type = mtype
+            cache_info = self.cache.get(meta)
+        if not cache_info:
+            # 缓存没有或者强制不使用缓存
+            if doubanid:
+                # 直接查询详情
+                info = self.douban_info(doubanid=doubanid, mtype=mtype or meta.type)
+            elif meta:
+                if meta.begin_season:
+                    logger.info(f"正在识别 {meta.name} 第{meta.begin_season}季 ...")
+                else:
+                    logger.info(f"正在识别 {meta.name} ...")
+                # 匹配豆瓣信息
+                match_info = self.match_doubaninfo(name=meta.name,
+                                                   mtype=mtype or meta.type,
+                                                   year=meta.year,
+                                                   season=meta.begin_season)
+                if match_info:
+                    # 匹配到豆瓣信息
+                    info = self.douban_info(
+                        doubanid=match_info.get("id"),
+                        mtype=mtype or meta.type
+                    )
+                else:
+                    logger.info(f"{meta.name if meta else doubanid} 未匹配到豆瓣媒体信息")
+                    return None
+            else:
+                logger.error("识别媒体信息时未提供元数据或豆瓣ID")
+                return None
+            # 保存到缓存
+            if meta:
+                self.cache.update(meta, info)
+        else:
+            # 使用缓存信息
+            if cache_info.get("title"):
+                logger.info(f"{meta.name} 使用豆瓣识别缓存：{cache_info.get('title')}")
+                info = self.douban_info(mtype=cache_info.get("type"),
+                                        doubanid=cache_info.get("id"))
+            else:
+                logger.info(f"{meta.name} 使用豆瓣识别缓存：无法识别")
+                info = None
+
+        if info:
+            # 赋值TMDB信息并返回
+            mediainfo = MediaInfo(douban_info=info)
+            if meta:
+                logger.info(f"{meta.name} 豆瓣识别结果：{mediainfo.type.value} "
+                            f"{mediainfo.title_year} "
+                            f"{mediainfo.douban_id}")
+            else:
+                logger.info(f"{doubanid} 豆瓣识别结果：{mediainfo.type.value} "
+                            f"{mediainfo.title_year}")
+            return mediainfo
+        else:
+            logger.info(f"{meta.name if meta else doubanid} 未匹配到豆瓣媒体信息")
+
+        return None
+
+    def douban_info(self, doubanid: str, mtype: MediaType = None) -> Optional[dict]:
         """
         获取豆瓣信息
         :param doubanid: 豆瓣ID
+        :param mtype:    媒体类型
         :return: 豆瓣信息
         """
         """
@@ -300,22 +380,40 @@ class DoubanModule(_ModuleBase):
           "interest_cmt_earlier_tip_desc": "该短评的发布时间早于公开上映时间，作者可能通过其他渠道提前观看，请谨慎参考。其评分将不计入总评分。"
         }
         """
+
+        def __douban_tv():
+            """
+            获取豆瓣剧集信息
+            """
+            info = self.doubanapi.tv_detail(doubanid)
+            if info:
+                celebrities = self.doubanapi.tv_celebrities(doubanid)
+                if celebrities:
+                    info["directors"] = celebrities.get("directors")
+                    info["actors"] = celebrities.get("actors")
+            return info
+
+        def __douban_movie():
+            """
+            获取豆瓣电影信息
+            """
+            info = self.doubanapi.movie_detail(doubanid)
+            if info:
+                celebrities = self.doubanapi.movie_celebrities(doubanid)
+                if celebrities:
+                    info["directors"] = celebrities.get("directors")
+                    info["actors"] = celebrities.get("actors")
+            return info
+
         if not doubanid:
             return None
         logger.info(f"开始获取豆瓣信息：{doubanid} ...")
-        douban_info = self.doubanapi.movie_detail(doubanid)
-        if douban_info:
-            celebrities = self.doubanapi.movie_celebrities(doubanid)
-            if celebrities:
-                douban_info["directors"] = celebrities.get("directors")
-                douban_info["actors"] = celebrities.get("actors")
+        if mtype == MediaType.TV:
+            return __douban_tv()
+        elif mtype == MediaType.MOVIE:
+            return __douban_movie()
         else:
-            douban_info = self.doubanapi.tv_detail(doubanid)
-            celebrities = self.doubanapi.tv_celebrities(doubanid)
-            if douban_info and celebrities:
-                douban_info["directors"] = celebrities.get("directors")
-                douban_info["actors"] = celebrities.get("actors")
-        return douban_info
+            return __douban_movie() or __douban_tv()
 
     def douban_discover(self, mtype: MediaType, sort: str, tags: str,
                         page: int = 1, count: int = 30) -> Optional[List[dict]]:
@@ -407,12 +505,12 @@ class DoubanModule(_ModuleBase):
 
     @retry(Exception, 5, 3, 3, logger=logger)
     def match_doubaninfo(self, name: str, imdbid: str = None,
-                         mtype: str = None, year: str = None, season: int = None) -> dict:
+                         mtype: MediaType = None, year: str = None, season: int = None) -> dict:
         """
         搜索和匹配豆瓣信息
         :param name:  名称
         :param imdbid:  IMDB ID
-        :param mtype:  类型 电影/电视剧
+        :param mtype:  类型
         :param year:  年份
         :param season:  季号
         """
@@ -441,7 +539,7 @@ class DoubanModule(_ModuleBase):
             type_name = item_obj.get("type_name")
             if type_name not in [MediaType.TV.value, MediaType.MOVIE.value]:
                 continue
-            if mtype and mtype != type_name:
+            if mtype and mtype.value != type_name:
                 continue
             if mtype == MediaType.TV and not season:
                 season = 1
@@ -486,17 +584,20 @@ class DoubanModule(_ModuleBase):
             meta = MetaInfo(path.stem)
             if not meta.name:
                 return
-            # 根据名称查询豆瓣数据
-            doubaninfo = self.match_doubaninfo(name=mediainfo.title,
-                                               imdbid=mediainfo.imdb_id,
-                                               mtype=mediainfo.type.value,
-                                               year=mediainfo.year,
-                                               season=meta.begin_season)
-            if not doubaninfo:
-                logger.warn(f"未找到 {mediainfo.title} 的豆瓣信息")
-                return
             # 查询豆瓣详情
-            doubaninfo = self.douban_info(doubaninfo.get("id"))
+            if not mediainfo.douban_id:
+                # 根据名称查询豆瓣数据
+                doubaninfo = self.match_doubaninfo(name=mediainfo.title,
+                                                   imdbid=mediainfo.imdb_id,
+                                                   mtype=mediainfo.type,
+                                                   year=mediainfo.year)
+                if not doubaninfo:
+                    logger.warn(f"未找到 {mediainfo.title} 的豆瓣信息")
+                    return
+                doubaninfo = self.douban_info(doubanid=doubaninfo.get("id"), mtype=mediainfo.type)
+            else:
+                doubaninfo = self.douban_info(doubanid=mediainfo.douban_id,
+                                              mtype=mediainfo.type)
             # 刮削路径
             scrape_path = path / path.name
             self.scraper.gen_scraper_files(meta=meta,
@@ -513,17 +614,21 @@ class DoubanModule(_ModuleBase):
                     meta = MetaInfo(file.stem)
                     if not meta.name:
                         continue
-                    # 根据名称查询豆瓣数据
-                    doubaninfo = self.match_doubaninfo(name=mediainfo.title,
-                                                       imdbid=mediainfo.imdb_id,
-                                                       mtype=mediainfo.type.value,
-                                                       year=mediainfo.year,
-                                                       season=meta.begin_season)
-                    if not doubaninfo:
-                        logger.warn(f"未找到 {mediainfo.title} 的豆瓣信息")
-                        break
-                    # 查询豆瓣详情
-                    doubaninfo = self.douban_info(doubaninfo.get("id"))
+                    if not mediainfo.douban_id:
+                        # 根据名称查询豆瓣数据
+                        doubaninfo = self.match_doubaninfo(name=mediainfo.title,
+                                                           imdbid=mediainfo.imdb_id,
+                                                           mtype=mediainfo.type,
+                                                           year=mediainfo.year,
+                                                           season=meta.begin_season)
+                        if not doubaninfo:
+                            logger.warn(f"未找到 {mediainfo.title} 的豆瓣信息")
+                            break
+                        # 查询豆瓣详情
+                        doubaninfo = self.douban_info(doubanid=doubaninfo.get("id"), mtype=mediainfo.type)
+                    else:
+                        doubaninfo = self.douban_info(doubanid=mediainfo.douban_id,
+                                                      mtype=mediainfo.type)
                     # 刮削
                     self.scraper.gen_scraper_files(meta=meta,
                                                    mediainfo=MediaInfo(douban_info=doubaninfo),
@@ -532,3 +637,10 @@ class DoubanModule(_ModuleBase):
                 except Exception as e:
                     logger.error(f"刮削文件 {file} 失败，原因：{str(e)}")
         logger.info(f"{path} 刮削完成")
+
+    def clear_cache(self):
+        """
+        清除缓存
+        """
+        self.doubanapi.clear_cache()
+        self.cache.clear()

app/modules/douban/apiv2.py

@@ -427,6 +427,12 @@ class DoubanApi(metaclass=Singleton):
         return self.__invoke(self._urls["doulist_items"] % subject_id,
                              start=start, count=count, _ts=ts)
 
+    def clear_cache(self):
+        """
+        清空LRU缓存
+        """
+        self.__invoke.cache_clear()
+
     def __del__(self):
         if self._session:
             self._session.close()

app/modules/douban/douban_cache.py

@@ -0,0 +1,228 @@
+import pickle
+import random
+import time
+from pathlib import Path
+from threading import RLock
+from typing import Optional
+
+from app.core.config import settings
+from app.core.meta import MetaBase
+from app.utils.singleton import Singleton
+from app.schemas.types import MediaType
+
+lock = RLock()
+
+CACHE_EXPIRE_TIMESTAMP_STR = "cache_expire_timestamp"
+EXPIRE_TIMESTAMP = settings.CACHE_CONF.get('meta')
+
+
+class DoubanCache(metaclass=Singleton):
+    """
+    豆瓣缓存数据
+    {
+        "id": '',
+        "title": '',
+        "year": '',
+        "type": MediaType
+    }
+    """
+    _meta_data: dict = {}
+    # 缓存文件路径
+    _meta_path: Path = None
+    # TMDB缓存过期
+    _tmdb_cache_expire: bool = True
+
+    def __init__(self):
+        self._meta_path = settings.TEMP_PATH / "__douban_cache__"
+        self._meta_data = self.__load(self._meta_path)
+
+    def clear(self):
+        """
+        清空所有TMDB缓存
+        """
+        with lock:
+            self._meta_data = {}
+
+    @staticmethod
+    def __get_key(meta: MetaBase) -> str:
+        """
+        获取缓存KEY
+        """
+        return f"[{meta.type.value if meta.type else '未知'}]{meta.name}-{meta.year}-{meta.begin_season}"
+
+    def get(self, meta: MetaBase):
+        """
+        根据KEY值获取缓存值
+        """
+        key = self.__get_key(meta)
+        with lock:
+            info: dict = self._meta_data.get(key)
+            if info:
+                expire = info.get(CACHE_EXPIRE_TIMESTAMP_STR)
+                if not expire or int(time.time()) < expire:
+                    info[CACHE_EXPIRE_TIMESTAMP_STR] = int(time.time()) + EXPIRE_TIMESTAMP
+                    self._meta_data[key] = info
+                elif expire and self._tmdb_cache_expire:
+                    self.delete(key)
+            return info or {}
+
+    def delete(self, key: str) -> dict:
+        """
+        删除缓存信息
+        @param key: 缓存key
+        @return: 被删除的缓存内容
+        """
+        with lock:
+            return self._meta_data.pop(key, None)
+
+    def delete_by_doubanid(self, doubanid: str) -> None:
+        """
+        清空对应豆瓣ID的所有缓存记录，以强制更新TMDB中最新的数据
+        """
+        for key in list(self._meta_data):
+            if self._meta_data.get(key, {}).get("id") == doubanid:
+                with lock:
+                    self._meta_data.pop(key)
+
+    def delete_unknown(self) -> None:
+        """
+        清除未识别的缓存记录，以便重新搜索TMDB
+        """
+        for key in list(self._meta_data):
+            if self._meta_data.get(key, {}).get("id") == "0":
+                with lock:
+                    self._meta_data.pop(key)
+
+    def modify(self, key: str, title: str) -> dict:
+        """
+        删除缓存信息
+        @param key: 缓存key
+        @param title: 标题
+        @return: 被修改后缓存内容
+        """
+        with lock:
+            if self._meta_data.get(key):
+                self._meta_data[key]['title'] = title
+                self._meta_data[key][CACHE_EXPIRE_TIMESTAMP_STR] = int(time.time()) + EXPIRE_TIMESTAMP
+            return self._meta_data.get(key)
+
+    @staticmethod
+    def __load(path: Path) -> dict:
+        """
+        从文件中加载缓存
+        """
+        try:
+            if path.exists():
+                with open(path, 'rb') as f:
+                    data = pickle.load(f)
+                return data
+            return {}
+        except Exception as e:
+            print(str(e))
+            return {}
+
+    def update(self, meta: MetaBase, info: dict) -> None:
+        """
+        新增或更新缓存条目
+        """
+        with lock:
+            if info:
+                # 缓存标题
+                cache_title = info.get("title") \
+                    if info.get("media_type") == MediaType.MOVIE else info.get("name")
+                # 缓存年份
+                cache_year = info.get('release_date') \
+                    if info.get("media_type") == MediaType.MOVIE else info.get('first_air_date')
+                if cache_year:
+                    cache_year = cache_year[:4]
+                # 类型
+                if isinstance(info.get('media_type'), MediaType):
+                    mtype = info.get('media_type')
+                else:
+                    mtype = MediaType.MOVIE if info.get("type") == "movie" else MediaType.TV
+                # 海报
+                poster_path = info.get("pic", {}).get("large")
+                if not poster_path and info.get("cover_url"):
+                    poster_path = info.get("cover_url")
+                if not poster_path and info.get("cover"):
+                    poster_path = info.get("cover").get("url")
+                self._meta_data[self.__get_key(meta)] = {
+                        "id": info.get("id"),
+                        "type": mtype,
+                        "year": cache_year,
+                        "title": cache_title,
+                        "poster_path": poster_path,
+                        CACHE_EXPIRE_TIMESTAMP_STR: int(time.time()) + EXPIRE_TIMESTAMP
+                    }
+            elif info is not None:
+                # None时不缓存，此时代表网络错误，允许重复请求
+                self._meta_data[self.__get_key(meta)] = {'id': "0"}
+
+    def save(self, force: bool = False) -> None:
+        """
+        保存缓存数据到文件
+        """
+
+        meta_data = self.__load(self._meta_path)
+        new_meta_data = {k: v for k, v in self._meta_data.items() if v.get("id")}
+
+        if not force \
+                and not self._random_sample(new_meta_data) \
+                and meta_data.keys() == new_meta_data.keys():
+            return
+
+        with open(self._meta_path, 'wb') as f:
+            pickle.dump(new_meta_data, f, pickle.HIGHEST_PROTOCOL)
+
+    def _random_sample(self, new_meta_data: dict) -> bool:
+        """
+        采样分析是否需要保存
+        """
+        ret = False
+        if len(new_meta_data) < 25:
+            keys = list(new_meta_data.keys())
+            for k in keys:
+                info = new_meta_data.get(k)
+                expire = info.get(CACHE_EXPIRE_TIMESTAMP_STR)
+                if not expire:
+                    ret = True
+                    info[CACHE_EXPIRE_TIMESTAMP_STR] = int(time.time()) + EXPIRE_TIMESTAMP
+                elif int(time.time()) >= expire:
+                    ret = True
+                    if self._tmdb_cache_expire:
+                        new_meta_data.pop(k)
+        else:
+            count = 0
+            keys = random.sample(sorted(new_meta_data.keys()), 25)
+            for k in keys:
+                info = new_meta_data.get(k)
+                expire = info.get(CACHE_EXPIRE_TIMESTAMP_STR)
+                if not expire:
+                    ret = True
+                    info[CACHE_EXPIRE_TIMESTAMP_STR] = int(time.time()) + EXPIRE_TIMESTAMP
+                elif int(time.time()) >= expire:
+                    ret = True
+                    if self._tmdb_cache_expire:
+                        new_meta_data.pop(k)
+                        count += 1
+            if count >= 5:
+                ret |= self._random_sample(new_meta_data)
+        return ret
+
+    def get_title(self, key: str) -> Optional[str]:
+        """
+        获取缓存的标题
+        """
+        cache_media_info = self._meta_data.get(key)
+        if not cache_media_info or not cache_media_info.get("id"):
+            return None
+        return cache_media_info.get("title")
+
+    def set_title(self, key: str, cn_title: str) -> None:
+        """
+        重新设置缓存标题
+        """
+        cache_media_info = self._meta_data.get(key)
+        if not cache_media_info:
+            return
+        self._meta_data[key]['title'] = cn_title

app/modules/fanart/__init__.py

@@ -326,17 +326,19 @@ class FanartModule(_ModuleBase):
         :param mediainfo:  识别的媒体信息
         :return: 更新后的媒体信息
         """
+        if not mediainfo.tmdb_id and not mediainfo.tvdb_id:
+            return None
         if mediainfo.type == MediaType.MOVIE:
             result = self.__request_fanart(mediainfo.type, mediainfo.tmdb_id)
         else:
             if mediainfo.tvdb_id:
                 result = self.__request_fanart(mediainfo.type, mediainfo.tvdb_id)
             else:
-                logger.info(f"{mediainfo.title_year} 没有tvdbid，无法获取Fanart图片")
-                return
+                logger.info(f"{mediainfo.title_year} 没有tvdbid，无法获取fanart图片")
+                return None
         if not result or result.get('status') == 'error':
-            logger.warn(f"没有获取到 {mediainfo.title_year} 的Fanart图片数据")
-            return
+            logger.warn(f"没有获取到 {mediainfo.title_year} 的fanart图片数据")
+            return None
         # 获取所有图片
         for name, images in result.items():
             if not images:

app/modules/themoviedb/__init__.py

@@ -43,7 +43,8 @@ class TheMovieDbModule(_ModuleBase):
 
     def recognize_media(self, meta: MetaBase = None,
                         mtype: MediaType = None,
-                        tmdbid: int = None) -> Optional[MediaInfo]:
+                        tmdbid: int = None,
+                        **kwargs) -> Optional[MediaInfo]:
         """
         识别媒体信息
         :param meta:     识别的元数据
@@ -51,6 +52,9 @@ class TheMovieDbModule(_ModuleBase):
         :param tmdbid:   tmdbid
         :return: 识别的媒体信息，包括剧集信息
         """
+        if settings.RECOGNIZE_SOURCE != "themoviedb":
+            return None
+
         if not meta:
             cache_info = {}
         else:
@@ -112,11 +116,11 @@ class TheMovieDbModule(_ModuleBase):
         else:
             # 使用缓存信息
             if cache_info.get("title"):
-                logger.info(f"{meta.name} 使用识别缓存：{cache_info.get('title')}")
+                logger.info(f"{meta.name} 使用TMDB识别缓存：{cache_info.get('title')}")
                 info = self.tmdb.get_info(mtype=cache_info.get("type"),
                                           tmdbid=cache_info.get("id"))
             else:
-                logger.info(f"{meta.name} 使用识别缓存：无法识别")
+                logger.info(f"{meta.name} 使用TMDB识别缓存：无法识别")
                 info = None
 
         if info:
@@ -129,11 +133,11 @@ class TheMovieDbModule(_ModuleBase):
             mediainfo = MediaInfo(tmdb_info=info)
             mediainfo.set_category(cat)
             if meta:
-                logger.info(f"{meta.name} 识别结果：{mediainfo.type.value} "
+                logger.info(f"{meta.name} TMDB识别结果：{mediainfo.type.value} "
                             f"{mediainfo.title_year} "
                             f"{mediainfo.tmdb_id}")
             else:
-                logger.info(f"{tmdbid} 识别结果：{mediainfo.type.value} "
+                logger.info(f"{tmdbid} TMDB识别结果：{mediainfo.type.value} "
                             f"{mediainfo.title_year}")
 
             # 补充剧集年份
@@ -143,10 +147,31 @@ class TheMovieDbModule(_ModuleBase):
                     mediainfo.season_years = episode_years
             return mediainfo
         else:
-            logger.info(f"{meta.name if meta else tmdbid} 未匹配到媒体信息")
+            logger.info(f"{meta.name if meta else tmdbid} 未匹配到TMDB媒体信息")
 
         return None
 
+    def match_doubaninfo(self, name: str, mtype: MediaType = None,
+                         year: str = None, season: int = None) -> dict:
+        """
+        搜索和匹配TMDB信息
+        :param name:  名称
+        :param mtype:  类型
+        :param year:  年份
+        :param season:  季号
+        """
+        # 搜索
+        logger.info(f"开始使用 名称：{name}、年份：{year} 匹配TMDB信息 ...")
+        info = self.tmdb.match(name=name,
+                               year=year,
+                               mtype=mtype,
+                               season_year=year,
+                               season_number=season)
+        if info and not info.get("genres"):
+            info = self.tmdb.get_info(mtype=info.get("media_type"),
+                                      tmdbid=info.get("id"))
+        return info
+
     def tmdb_info(self, tmdbid: int, mtype: MediaType) -> Optional[dict]:
         """
         获取TMDB信息

tests/test_recognize.py

@@ -15,7 +15,7 @@ class RecognizeTest(TestCase):
         pass
 
     def test_recognize(self):
-        result = MediaChain().recognize_by_title(title="我和我的祖国 2019")
-        self.assertEqual(result.media_info.tmdb_id, 612845)
-        exists = DownloadChain().get_no_exists_info(MetaInfo("我和我的祖国 2019"), result.media_info)
+        media_info = MediaChain().recognize_by_title(title="我和我的祖国 2019")
+        self.assertEqual(media_info.tmdb_id, 612845)
+        exists = DownloadChain().get_no_exists_info(MetaInfo("我和我的祖国 2019"), media_info)
         self.assertTrue(exists[0])