Merge pull request #4561 from jxxghp/cursor/move-media-files-to-season-directory-6ee0

Co-authored-by: jxxghp <jxxghp@163.com>

app/chain/media.py

@@ -19,7 +19,6 @@ from app.utils.string import StringUtils
 
 recognize_lock = Lock()
 scraping_lock = Lock()
-scraping_files = []
 
 
 class MediaChain(ChainBase):
@@ -35,25 +34,25 @@ class MediaChain(ChainBase):
         switchs = SystemConfigOper().get(SystemConfigKey.ScrapingSwitchs) or {}
         # 默认配置
         default_switchs = {
-            'movie_nfo': True,          # 电影NFO
-            'movie_poster': True,       # 电影海报
-            'movie_backdrop': True,     # 电影背景图
-            'movie_logo': True,         # 电影Logo
-            'movie_disc': True,         # 电影光盘图
-            'movie_banner': True,       # 电影横幅图
-            'movie_thumb': True,        # 电影缩略图
-            'tv_nfo': True,            # 电视剧NFO
-            'tv_poster': True,         # 电视剧海报
-            'tv_backdrop': True,       # 电视剧背景图
-            'tv_banner': True,         # 电视剧横幅图
-            'tv_logo': True,           # 电视剧Logo
-            'tv_thumb': True,          # 电视剧缩略图
-            'season_nfo': True,        # 季NFO
-            'season_poster': True,     # 季海报
-            'season_banner': True,     # 季横幅图
-            'season_thumb': True,      # 季缩略图
-            'episode_nfo': True,       # 集NFO
-            'episode_thumb': True      # 集缩略图
+            'movie_nfo': True,  # 电影NFO
+            'movie_poster': True,  # 电影海报
+            'movie_backdrop': True,  # 电影背景图
+            'movie_logo': True,  # 电影Logo
+            'movie_disc': True,  # 电影光盘图
+            'movie_banner': True,  # 电影横幅图
+            'movie_thumb': True,  # 电影缩略图
+            'tv_nfo': True,  # 电视剧NFO
+            'tv_poster': True,  # 电视剧海报
+            'tv_backdrop': True,  # 电视剧背景图
+            'tv_banner': True,  # 电视剧横幅图
+            'tv_logo': True,  # 电视剧Logo
+            'tv_thumb': True,  # 电视剧缩略图
+            'season_nfo': True,  # 季NFO
+            'season_poster': True,  # 季海报
+            'season_banner': True,  # 季横幅图
+            'season_thumb': True,  # 季缩略图
+            'episode_nfo': True,  # 集NFO
+            'episode_thumb': True  # 集缩略图
         }
         # 合并用户配置和默认配置
         for key, default_value in default_switchs.items():
@@ -344,23 +343,49 @@ class MediaChain(ChainBase):
             return
         event_data = event.event_data or {}
         fileitem: FileItem = event_data.get("fileitem")
+        file_list: List[str] = event_data.get("file_list", [])
         meta: MetaBase = event_data.get("meta")
         mediainfo: MediaInfo = event_data.get("mediainfo")
         overwrite = event_data.get("overwrite", False)
         if not fileitem:
             return
+
         # 刮削锁
         with scraping_lock:
-            if fileitem.path in scraping_files:
+            # 检查文件项是否存在
+            storagechain = StorageChain()
+            if not storagechain.get_item(fileitem):
+                logger.warn(f"文件项不存在：{fileitem.path}")
                 return
-            scraping_files.append(fileitem.path)
-        try:
-            # 执行刮削
-            self.scrape_metadata(fileitem=fileitem, meta=meta, mediainfo=mediainfo, overwrite=overwrite)
-        finally:
-            # 释放锁
-            with scraping_lock:
-                scraping_files.remove(fileitem.path)
+            # 检查是否为目录
+            if fileitem.type == "file":
+                # 单个文件刮削
+                self.scrape_metadata(fileitem=fileitem,
+                                     mediainfo=mediainfo,
+                                     init_folder=False,
+                                     parent=storagechain.get_parent_item(fileitem),
+                                     overwrite=overwrite)
+            else:
+                # 检查目的目录下是否已经有nfo刮削文件
+                sub_files = storagechain.list_files(fileitem)
+                if any(f.name.endswith('.nfo') for f in sub_files):
+                    logger.info(f"目录 {fileitem.path} 已有NFO文件，开始增量刮削...")
+                    for file_path in file_list:
+                        file_item = storagechain.get_file_item(storage=fileitem.storage,
+                                                              path=Path(file_path))
+                        if file_item:
+                            # 对于电视剧文件，应该保存到与视频文件相同的目录
+                            # 而不是电视剧根目录
+                            self.scrape_metadata(fileitem=file_item,
+                                                 mediainfo=mediainfo,
+                                                 init_folder=False,
+                                                 parent=None,  # 让函数内部自动获取正确的父目录
+                                                 overwrite=overwrite)
+                else:
+                    # 执行全量刮削
+                    logger.info(f"开始全量刮削目录 {fileitem.path} ...")
+                    self.scrape_metadata(fileitem=fileitem, meta=meta, init_folder=True,
+                                         mediainfo=mediainfo, overwrite=overwrite)
 
     def scrape_metadata(self, fileitem: schemas.FileItem,
                         meta: MetaBase = None, mediainfo: MediaInfo = None,
@@ -436,6 +461,9 @@ class MediaChain(ChainBase):
                 logger.error(f"{_url} 图片下载失败：{str(err)}！")
             return None
 
+        if not fileitem:
+            return
+
         # 当前文件路径
         filepath = Path(fileitem.path)
         if fileitem.type == "file" \
@@ -464,6 +492,8 @@ class MediaChain(ChainBase):
                         movie_nfo = self.metadata_nfo(meta=meta, mediainfo=mediainfo)
                         if movie_nfo:
                             # 保存或上传nfo文件到上级目录
+                            if not parent:
+                                parent = storagechain.get_parent_item(fileitem)
                             __save_file(_fileitem=parent, _path=nfo_path, _content=movie_nfo)
                         else:
                             logger.warn(f"{filepath.name} nfo文件生成失败！")
@@ -494,8 +524,9 @@ class MediaChain(ChainBase):
                     files = __list_files(_fileitem=fileitem)
                     for file in files:
                         self.scrape_metadata(fileitem=file,
-                                             meta=meta, mediainfo=mediainfo,
-                                             init_folder=False, parent=fileitem,
+                                             mediainfo=mediainfo,
+                                             init_folder=False,
+                                             parent=fileitem,
                                              overwrite=overwrite)
                 # 生成目录内图片文件
                 if init_folder:
@@ -587,11 +618,11 @@ class MediaChain(ChainBase):
                 else:
                     logger.info("集缩略图刮削已关闭，跳过")
             else:
-                # 当前为目录，处理目录内的文件
+                # 当前为电视剧目录，处理目录内的文件
                 files = __list_files(_fileitem=fileitem)
                 for file in files:
                     self.scrape_metadata(fileitem=file,
-                                         meta=meta, mediainfo=mediainfo,
+                                         mediainfo=mediainfo,
                                          parent=fileitem if file.type == "file" else None,
                                          init_folder=True if file.type == "dir" else False,
                                          overwrite=overwrite)
@@ -659,7 +690,8 @@ class MediaChain(ChainBase):
                                         # 只下载当前刮削季的图片
                                         image_season = "00" if "specials" in image_name else image_name[6:8]
                                         if image_season != str(season_meta.begin_season).rjust(2, '0'):
-                                            logger.info(f"当前刮削季为：{season_meta.begin_season}，跳过文件：{image_path}")
+                                            logger.info(
+                                                f"当前刮削季为：{season_meta.begin_season}，跳过文件：{image_path}")
                                             continue
                                         if overwrite or not storagechain.get_file_item(storage=fileitem.storage,
                                                                                        path=image_path):

app/chain/site.py

@@ -271,16 +271,20 @@ class SiteChain(ChainBase):
             logger.error(f"获取站点页面失败：{url}")
             return favicon_url, None
         html = etree.HTML(html_text)
-        if StringUtils.is_valid_html_element(html):
-            fav_link = html.xpath('//head/link[contains(@rel, "icon")]/@href')
-            if fav_link:
-                favicon_url = urljoin(url, fav_link[0])
+        try:
+            if StringUtils.is_valid_html_element(html):
+                fav_link = html.xpath('//head/link[contains(@rel, "icon")]/@href')
+                if fav_link:
+                    favicon_url = urljoin(url, fav_link[0])
 
-        res = RequestUtils(cookies=cookie, timeout=15, ua=ua).get_res(url=favicon_url)
-        if res:
-            return favicon_url, base64.b64encode(res.content).decode()
-        else:
-            logger.error(f"获取站点图标失败：{favicon_url}")
+            res = RequestUtils(cookies=cookie, timeout=15, ua=ua).get_res(url=favicon_url)
+            if res:
+                return favicon_url, base64.b64encode(res.content).decode()
+            else:
+                logger.error(f"获取站点图标失败：{favicon_url}")
+        finally:
+            if html is not None:
+                del html
         return favicon_url, None
 
     def sync_cookies(self, manual=False) -> Tuple[bool, str]:

app/chain/subscribe.py

@@ -39,6 +39,8 @@ class SubscribeChain(ChainBase):
     """
 
     _rlock = threading.RLock()
+    # 避免莫名原因导致长时间持有锁
+    _LOCK_TIMOUT = 3600 * 2
 
     def add(self, title: str, year: str,
             mtype: MediaType = None,
@@ -279,8 +281,15 @@ class SubscribeChain(ChainBase):
         :param manual: 是否手动搜索
         :return: 更新订阅状态为R或删除订阅
         """
-        with self._rlock:
-            logger.debug(f"search lock acquired at {datetime.now()}")
+        lock_acquired = False
+        try:
+            if lock_acquired := self._rlock.acquire(
+                blocking=True, timeout=self._LOCK_TIMOUT
+            ):
+                logger.debug(f"search lock acquired at {datetime.now()}")
+            else:
+                logger.warn("search上锁超时")
+
             subscribeoper = SubscribeOper()
             if sid:
                 subscribe = subscribeoper.get(sid)
@@ -437,8 +446,10 @@ class SubscribeChain(ChainBase):
             finally:
                 subscribes.clear()
                 del subscribes
-
-            logger.debug(f"search Lock released at {datetime.now()}")
+        finally:
+            if lock_acquired:
+                self._rlock.release()
+                logger.debug(f"search Lock released at {datetime.now()}")
 
         # 如果不是大内存模式，进行垃圾回收
         if not settings.BIG_MEMORY_MODE:
@@ -565,8 +576,14 @@ class SubscribeChain(ChainBase):
             logger.warn('没有缓存资源，无法匹配订阅')
             return
 
-        with self._rlock:
-            logger.debug(f"match lock acquired at {datetime.now()}")
+        lock_acquired = False
+        try:
+            if lock_acquired := self._rlock.acquire(
+                blocking=True, timeout=self._LOCK_TIMOUT
+            ):
+                logger.debug(f"match lock acquired at {datetime.now()}")
+            else:
+                logger.warn("match上锁超时")
 
             # 预识别所有未识别的种子
             processed_torrents: Dict[str, List[Context]] = {}
@@ -822,8 +839,10 @@ class SubscribeChain(ChainBase):
                 del processed_torrents
                 subscribes.clear()
                 del subscribes
-
-            logger.debug(f"match Lock released at {datetime.now()}")
+        finally:
+            if lock_acquired:
+                self._rlock.release()
+                logger.debug(f"match Lock released at {datetime.now()}")
 
     def check(self):
         """

app/chain/transfer.py

@@ -488,7 +488,9 @@ class TransferChain(ChainBase, metaclass=Singleton):
                     self.eventmanager.send_event(EventType.MetadataScrape, {
                         'meta': task.meta,
                         'mediainfo': task.mediainfo,
-                        'fileitem': transferinfo.target_diritem
+                        'fileitem': transferinfo.target_diritem,
+                        'file_list': transferinfo.file_list_new,
+                        'overwrite': False
                     })
 
                 # 移除已完成的任务

app/core/meta/streamingplatform.py

@@ -312,4 +312,3 @@ class StreamingPlatforms(metaclass=Singleton):
         if name is None:
             return False
         return name.upper() in self._lookup_cache
-

app/helper/cookie.py

@@ -96,53 +96,58 @@ class CookieHelper:
                 return None, None, "获取源码失败"
             # 查找用户名输入框
             html = etree.HTML(html_text)
-            username_xpath = None
-            for xpath in self._SITE_LOGIN_XPATH.get("username"):
-                if html.xpath(xpath):
-                    username_xpath = xpath
-                    break
-            if not username_xpath:
-                return None, None, "未找到用户名输入框"
-            # 查找密码输入框
-            password_xpath = None
-            for xpath in self._SITE_LOGIN_XPATH.get("password"):
-                if html.xpath(xpath):
-                    password_xpath = xpath
-                    break
-            if not password_xpath:
-                return None, None, "未找到密码输入框"
-            # 处理二步验证码
-            otp_code = TwoFactorAuth(two_step_code).get_code()
-            # 查找二步验证码输入框
-            twostep_xpath = None
-            if otp_code:
-                for xpath in self._SITE_LOGIN_XPATH.get("twostep"):
+            try:
+                username_xpath = None
+                for xpath in self._SITE_LOGIN_XPATH.get("username"):
                     if html.xpath(xpath):
-                        twostep_xpath = xpath
+                        username_xpath = xpath
                         break
-            # 查找验证码输入框
-            captcha_xpath = None
-            for xpath in self._SITE_LOGIN_XPATH.get("captcha"):
-                if html.xpath(xpath):
-                    captcha_xpath = xpath
-                    break
-            # 查找验证码图片
-            captcha_img_url = None
-            if captcha_xpath:
-                for xpath in self._SITE_LOGIN_XPATH.get("captcha_img"):
+                if not username_xpath:
+                    return None, None, "未找到用户名输入框"
+                # 查找密码输入框
+                password_xpath = None
+                for xpath in self._SITE_LOGIN_XPATH.get("password"):
                     if html.xpath(xpath):
-                        captcha_img_url = html.xpath(xpath)[0]
+                        password_xpath = xpath
                         break
-                if not captcha_img_url:
-                    return None, None, "未找到验证码图片"
-            # 查找登录按钮
-            submit_xpath = None
-            for xpath in self._SITE_LOGIN_XPATH.get("submit"):
-                if html.xpath(xpath):
-                    submit_xpath = xpath
-                    break
-            if not submit_xpath:
-                return None, None, "未找到登录按钮"
+                if not password_xpath:
+                    return None, None, "未找到密码输入框"
+                # 处理二步验证码
+                otp_code = TwoFactorAuth(two_step_code).get_code()
+                # 查找二步验证码输入框
+                twostep_xpath = None
+                if otp_code:
+                    for xpath in self._SITE_LOGIN_XPATH.get("twostep"):
+                        if html.xpath(xpath):
+                            twostep_xpath = xpath
+                            break
+                # 查找验证码输入框
+                captcha_xpath = None
+                for xpath in self._SITE_LOGIN_XPATH.get("captcha"):
+                    if html.xpath(xpath):
+                        captcha_xpath = xpath
+                        break
+                # 查找验证码图片
+                captcha_img_url = None
+                if captcha_xpath:
+                    for xpath in self._SITE_LOGIN_XPATH.get("captcha_img"):
+                        if html.xpath(xpath):
+                            captcha_img_url = html.xpath(xpath)[0]
+                            break
+                    if not captcha_img_url:
+                        return None, None, "未找到验证码图片"
+                # 查找登录按钮
+                submit_xpath = None
+                for xpath in self._SITE_LOGIN_XPATH.get("submit"):
+                    if html.xpath(xpath):
+                        submit_xpath = xpath
+                        break
+                if not submit_xpath:
+                    return None, None, "未找到登录按钮"
+            finally:
+                if html is not None:
+                    del html
+
             # 点击登录按钮
             try:
                 # 等待登录按钮准备好
@@ -185,19 +190,23 @@ class CookieHelper:
                 if not otp_code:
                     return None, None, "需要二次验证码"
                 html = etree.HTML(page.content())
-                for xpath in self._SITE_LOGIN_XPATH.get("twostep"):
-                    if html.xpath(xpath):
-                        try:
-                            # 刷新一下 2fa code
-                            otp_code = TwoFactorAuth(two_step_code).get_code()
-                            page.fill(xpath, otp_code)
-                            # 登录按钮 xpath 理论上相同，不再重复查找
-                            page.click(submit_xpath)
-                            page.wait_for_load_state("networkidle", timeout=30 * 1000)
-                        except Exception as e:
-                            logger.error(f"二次验证码输入失败：{str(e)}")
-                            return None, None, f"二次验证码输入失败：{str(e)}"
-                        break
+                try:
+                    for xpath in self._SITE_LOGIN_XPATH.get("twostep"):
+                        if html.xpath(xpath):
+                            try:
+                                # 刷新一下 2fa code
+                                otp_code = TwoFactorAuth(two_step_code).get_code()
+                                page.fill(xpath, otp_code)
+                                # 登录按钮 xpath 理论上相同，不再重复查找
+                                page.click(submit_xpath)
+                                page.wait_for_load_state("networkidle", timeout=30 * 1000)
+                            except Exception as e:
+                                logger.error(f"二次验证码输入失败：{str(e)}")
+                                return None, None, f"二次验证码输入失败：{str(e)}"
+                            break
+                finally:
+                    if html is not None:
+                        del html
             # 登录后的源码
             html_text = page.content()
             if not html_text:

app/helper/resource.py

@@ -8,6 +8,7 @@ from app.log import logger
 from app.utils.http import RequestUtils
 from app.utils.string import StringUtils
 from app.utils.system import SystemUtils
+from version import APP_VERSION
 
 
 class ResourceHelper:
@@ -58,15 +59,15 @@ class ResourceHelper:
                         if rtype == "auth":
                             # 站点认证资源
                             local_version = SitesHelper().auth_version
-                            # 阻断v2.3.0以下的版本直接更新，避免无限重启
+                            # 阻断站点认证资源v2.3.0以下的版本直接更新，避免无限重启
                             if StringUtils.compare_version(local_version, "<", "2.3.0"):
                                 continue
+                            # 阻断主程序版本v2.6.3以下的版本直接更新，避免搜索异常
+                            if StringUtils.compare_version(APP_VERSION, "<", "2.6.3"):
+                                continue
                         elif rtype == "sites":
                             # 站点索引资源
                             local_version = SitesHelper().indexer_version
-                            # 阻断v2.0.0以下的版本直接更新，避免无限重启
-                            if StringUtils.compare_version(local_version, "<", "2.0.0"):
-                                continue
                         else:
                             continue
                         if StringUtils.compare_version(version, ">", local_version):
@@ -84,6 +85,8 @@ class ResourceHelper:
                         elif not r:
                             return None, "连接仓库失败"
                         files_info = r.json()
+                        # 下载资源文件
+                        success = True
                         for item in files_info:
                             save_path = need_updates.get(item.get("name"))
                             if not save_path:
@@ -96,16 +99,23 @@ class ResourceHelper:
                                                    timeout=180).get_res(download_url)
                                 if not res:
                                     logger.error(f"文件 {item.get('name')} 下载失败！")
+                                    success = False
+                                    break
                                 elif res.status_code != 200:
                                     logger.error(f"下载文件 {item.get('name')} 失败：{res.status_code} - {res.reason}")
+                                    success = False
+                                    break
                                 # 创建插件文件夹
                                 file_path = self._base_dir / save_path / item.get("name")
                                 if not file_path.parent.exists():
                                     file_path.parent.mkdir(parents=True, exist_ok=True)
                                 # 写入文件
                                 file_path.write_bytes(res.content)
-                        logger.info("资源包更新完成，开始重启服务...")
-                        SystemHelper.restart()
+                        if success:
+                            logger.info("资源包更新完成，开始重启服务...")
+                            SystemHelper.restart()
+                        else:
+                            logger.warn("资源包更新失败，跳过升级！")
                     else:
                         logger.info("所有资源已最新，无需更新")
             except json.JSONDecodeError:

app/modules/filemanager/__init__.py

@@ -530,7 +530,14 @@ class FileManagerModule(_ModuleBase):
                                                     mediainfo=mediainfo)
             )
             # 计算重命名中的文件夹层数
-            rename_format_level = len(rename_format.split("/")) - 1
+            rename_list = rename_format.split("/")
+            rename_format_level = len(rename_list) - 1
+            for level, name in enumerate(rename_list):
+                # 处理特例，有的人重命名第一层是年份、分辨率
+                if "{{title}}" in name:
+                    # 找出含标题的这一层作为扫描路径
+                    rename_format_level -= level
+                    break
             # 取相对路径的第1层目录
             media_path = target_path.parents[rename_format_level - 1]
             if dir_path.is_relative_to(media_path):
@@ -562,9 +569,12 @@ class FileManagerModule(_ModuleBase):
         if not settings.LOCAL_EXISTS_SEARCH:
             return None
 
+        logger.debug(f"正在本地媒体库中查找 {mediainfo.title_year}...")
+
         # 检查媒体库
         fileitems = self.media_files(mediainfo)
         if not fileitems:
+            logger.debug(f"{mediainfo.title_year} 不在本地媒体库中")
             return None
 
         if mediainfo.type == MediaType.MOVIE:

app/modules/filemanager/storages/u115.py

@@ -219,8 +219,11 @@ class U115Pan(StorageBase, metaclass=Singleton):
 
         # 处理速率限制
         if resp.status_code == 429:
-            reset_time = int(resp.headers.get("X-RateLimit-Reset", 60))
-            time.sleep(reset_time + 5)
+            reset_time = 5 + int(resp.headers.get("X-RateLimit-Reset", 60))
+            logger.debug(
+                f"【115】{method} 请求 {endpoint} 限流，等待{reset_time}秒后重试"
+            )
+            time.sleep(reset_time)
             return self._request_api(method, endpoint, result_key, **kwargs)
 
         # 处理请求错误

app/modules/indexer/__init__.py

@@ -5,10 +5,11 @@ from app.core.config import settings
 from app.core.context import TorrentInfo
 from app.db.site_oper import SiteOper
 from app.helper.module import ModuleHelper
-from app.helper.sites import SitesHelper, SiteSpider
+from app.helper.sites import SitesHelper
 from app.log import logger
 from app.modules import _ModuleBase
 from app.modules.indexer.parser import SiteParserBase
+from app.modules.indexer.spider import SiteSpider
 from app.modules.indexer.spider.haidan import HaiDanSpider
 from app.modules.indexer.spider.hddolby import HddolbySpider
 from app.modules.indexer.spider.mtorrent import MTorrentSpider

app/modules/indexer/parser/discuz.py

@@ -14,15 +14,18 @@ class DiscuzUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
-
-        user_info = html.xpath('//a[contains(@href, "&uid=")]')
-        if user_info:
-            user_id_match = re.search(r"&uid=(\d+)", user_info[0].attrib['href'])
-            if user_id_match and user_id_match.group().strip():
-                self.userid = user_id_match.group(1)
-                self._torrent_seeding_page = f"forum.php?&mod=torrents&cat_5up=on"
-                self._user_detail_page = user_info[0].attrib['href']
-                self.username = user_info[0].text.strip()
+        try:
+            user_info = html.xpath('//a[contains(@href, "&uid=")]')
+            if user_info:
+                user_id_match = re.search(r"&uid=(\d+)", user_info[0].attrib['href'])
+                if user_id_match and user_id_match.group().strip():
+                    self.userid = user_id_match.group(1)
+                    self._torrent_seeding_page = f"forum.php?&mod=torrents&cat_5up=on"
+                    self._user_detail_page = user_info[0].attrib['href']
+                    self.username = user_info[0].text.strip()
+        finally:
+            if html is not None:
+                del html
 
     def _parse_site_page(self, html_text: str):
         pass
@@ -34,40 +37,44 @@ class DiscuzUserInfo(SiteParserBase):
         :return:
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        # 用户等级
-        user_levels_text = html.xpath('//a[contains(@href, "usergroup")]/text()')
-        if user_levels_text:
-            self.user_level = user_levels_text[-1].strip()
+            # 用户等级
+            user_levels_text = html.xpath('//a[contains(@href, "usergroup")]/text()')
+            if user_levels_text:
+                self.user_level = user_levels_text[-1].strip()
 
-        # 加入日期
-        join_at_text = html.xpath('//li[em[text()="注册时间"]]/text()')
-        if join_at_text:
-            self.join_at = StringUtils.unify_datetime_str(join_at_text[0].strip())
+            # 加入日期
+            join_at_text = html.xpath('//li[em[text()="注册时间"]]/text()')
+            if join_at_text:
+                self.join_at = StringUtils.unify_datetime_str(join_at_text[0].strip())
 
-        # 分享率
-        ratio_text = html.xpath('//li[contains(.//text(), "分享率")]//text()')
-        if ratio_text:
-            ratio_match = re.search(r"\(([\d,.]+)\)", ratio_text[0])
-            if ratio_match and ratio_match.group(1).strip():
-                self.bonus = StringUtils.str_float(ratio_match.group(1))
+            # 分享率
+            ratio_text = html.xpath('//li[contains(.//text(), "分享率")]//text()')
+            if ratio_text:
+                ratio_match = re.search(r"\(([\d,.]+)\)", ratio_text[0])
+                if ratio_match and ratio_match.group(1).strip():
+                    self.bonus = StringUtils.str_float(ratio_match.group(1))
 
-        # 积分
-        bouns_text = html.xpath('//li[em[text()="积分"]]/text()')
-        if bouns_text:
-            self.bonus = StringUtils.str_float(bouns_text[0].strip())
+            # 积分
+            bouns_text = html.xpath('//li[em[text()="积分"]]/text()')
+            if bouns_text:
+                self.bonus = StringUtils.str_float(bouns_text[0].strip())
 
-        # 上传
-        upload_text = html.xpath('//li[em[contains(text(),"上传量")]]/text()')
-        if upload_text:
-            self.upload = StringUtils.num_filesize(upload_text[0].strip().split('/')[-1])
+            # 上传
+            upload_text = html.xpath('//li[em[contains(text(),"上传量")]]/text()')
+            if upload_text:
+                self.upload = StringUtils.num_filesize(upload_text[0].strip().split('/')[-1])
 
-        # 下载
-        download_text = html.xpath('//li[em[contains(text(),"下载量")]]/text()')
-        if download_text:
-            self.download = StringUtils.num_filesize(download_text[0].strip().split('/')[-1])
+            # 下载
+            download_text = html.xpath('//li[em[contains(text(),"下载量")]]/text()')
+            if download_text:
+                self.download = StringUtils.num_filesize(download_text[0].strip().split('/')[-1])
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: bool = False) -> Optional[str]:
         """
@@ -77,44 +84,48 @@ class DiscuzUserInfo(SiteParserBase):
         :return: 下页地址
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        size_col = 3
-        seeders_col = 4
-        # 搜索size列
-        if html.xpath('//tr[position()=1]/td[.//img[@class="size"] and .//img[@alt="size"]]'):
-            size_col = len(html.xpath('//tr[position()=1]/td[.//img[@class="size"] '
-                                      'and .//img[@alt="size"]]/preceding-sibling::td')) + 1
-        # 搜索seeders列
-        if html.xpath('//tr[position()=1]/td[.//img[@class="seeders"] and .//img[@alt="seeders"]]'):
-            seeders_col = len(html.xpath('//tr[position()=1]/td[.//img[@class="seeders"] '
-                                         'and .//img[@alt="seeders"]]/preceding-sibling::td')) + 1
+            size_col = 3
+            seeders_col = 4
+            # 搜索size列
+            if html.xpath('//tr[position()=1]/td[.//img[@class="size"] and .//img[@alt="size"]]'):
+                size_col = len(html.xpath('//tr[position()=1]/td[.//img[@class="size"] '
+                                          'and .//img[@alt="size"]]/preceding-sibling::td')) + 1
+            # 搜索seeders列
+            if html.xpath('//tr[position()=1]/td[.//img[@class="seeders"] and .//img[@alt="seeders"]]'):
+                seeders_col = len(html.xpath('//tr[position()=1]/td[.//img[@class="seeders"] '
+                                             'and .//img[@alt="seeders"]]/preceding-sibling::td')) + 1
 
-        page_seeding = 0
-        page_seeding_size = 0
-        page_seeding_info = []
-        seeding_sizes = html.xpath(f'//tr[position()>1]/td[{size_col}]')
-        seeding_seeders = html.xpath(f'//tr[position()>1]/td[{seeders_col}]//text()')
-        if seeding_sizes and seeding_seeders:
-            page_seeding = len(seeding_sizes)
+            page_seeding = 0
+            page_seeding_size = 0
+            page_seeding_info = []
+            seeding_sizes = html.xpath(f'//tr[position()>1]/td[{size_col}]')
+            seeding_seeders = html.xpath(f'//tr[position()>1]/td[{seeders_col}]//text()')
+            if seeding_sizes and seeding_seeders:
+                page_seeding = len(seeding_sizes)
 
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = StringUtils.str_int(seeding_seeders[i])
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = StringUtils.str_int(seeding_seeders[i])
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        self.seeding += page_seeding
-        self.seeding_size += page_seeding_size
-        self.seeding_info.extend(page_seeding_info)
+            self.seeding += page_seeding
+            self.seeding_size += page_seeding_size
+            self.seeding_info.extend(page_seeding_info)
 
-        # 是否存在下页数据
-        next_page = None
-        next_page_text = html.xpath('//a[contains(.//text(), "下一页") or contains(.//text(), "下一頁")]/@href')
-        if next_page_text:
-            next_page = next_page_text[-1].strip()
+            # 是否存在下页数据
+            next_page = None
+            next_page_text = html.xpath('//a[contains(.//text(), "下一页") or contains(.//text(), "下一頁")]/@href')
+            if next_page_text:
+                next_page = next_page_text[-1].strip()
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 

app/modules/indexer/parser/file_list.py

@@ -24,10 +24,13 @@ class FileListSiteUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
-
-        ret = html.xpath(f'//a[contains(@href, "userdetails") and contains(@href, "{self.userid}")]//text()')
-        if ret:
-            self.username = str(ret[0])
+        try:
+            ret = html.xpath(f'//a[contains(@href, "userdetails") and contains(@href, "{self.userid}")]//text()')
+            if ret:
+                self.username = str(ret[0])
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_traffic_info(self, html_text: str):
         """
@@ -40,39 +43,41 @@ class FileListSiteUserInfo(SiteParserBase):
     def _parse_user_detail_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
+        try:
+            upload_html = html.xpath('//table//tr/td[text()="Uploaded"]/following-sibling::td//text()')
+            if upload_html:
+                self.upload = StringUtils.num_filesize(upload_html[0])
+            download_html = html.xpath('//table//tr/td[text()="Downloaded"]/following-sibling::td//text()')
+            if download_html:
+                self.download = StringUtils.num_filesize(download_html[0])
 
-        upload_html = html.xpath('//table//tr/td[text()="Uploaded"]/following-sibling::td//text()')
-        if upload_html:
-            self.upload = StringUtils.num_filesize(upload_html[0])
-        download_html = html.xpath('//table//tr/td[text()="Downloaded"]/following-sibling::td//text()')
-        if download_html:
-            self.download = StringUtils.num_filesize(download_html[0])
+            ratio_html = html.xpath('//table//tr/td[text()="Share ratio"]/following-sibling::td//text()')
+            if ratio_html:
+                share_ratio = StringUtils.str_float(ratio_html[0])
+            else:
+                share_ratio = 0
+            self.ratio = 0 if self.download == 0 else share_ratio
 
-        ratio_html = html.xpath('//table//tr/td[text()="Share ratio"]/following-sibling::td//text()')
-        if ratio_html:
-            share_ratio = StringUtils.str_float(ratio_html[0])
-        else:
-            share_ratio = 0
-        self.ratio = 0 if self.download == 0 else share_ratio
+            seed_html = html.xpath('//table//tr/td[text()="Seed bonus"]/following-sibling::td//text()')
+            if seed_html:
+                self.seeding = StringUtils.str_int(seed_html[1])
+                self.seeding_size = StringUtils.num_filesize(seed_html[3])
 
-        seed_html = html.xpath('//table//tr/td[text()="Seed bonus"]/following-sibling::td//text()')
-        if seed_html:
-            self.seeding = StringUtils.str_int(seed_html[1])
-            self.seeding_size = StringUtils.num_filesize(seed_html[3])
+            user_level_html = html.xpath('//table//tr/td[text()="Class"]/following-sibling::td//text()')
+            if user_level_html:
+                self.user_level = user_level_html[0].strip()
 
-        user_level_html = html.xpath('//table//tr/td[text()="Class"]/following-sibling::td//text()')
-        if user_level_html:
-            self.user_level = user_level_html[0].strip()
+            join_at_html = html.xpath('//table//tr/td[contains(text(), "Join")]/following-sibling::td//text()')
+            if join_at_html:
+                join_at = (join_at_html[0].split("("))[0].strip()
+                self.join_at = StringUtils.unify_datetime_str(join_at)
 
-        join_at_html = html.xpath('//table//tr/td[contains(text(), "Join")]/following-sibling::td//text()')
-        if join_at_html:
-            join_at = (join_at_html[0].split("("))[0].strip()
-            self.join_at = StringUtils.unify_datetime_str(join_at)
-
-        bonus_html = html.xpath('//a[contains(@href, "shop.php")]')
-        if bonus_html:
-            self.bonus = StringUtils.str_float(bonus_html[0].xpath("string(.)").strip())
-        pass
+            bonus_html = html.xpath('//a[contains(@href, "shop.php")]')
+            if bonus_html:
+                self.bonus = StringUtils.str_float(bonus_html[0].xpath("string(.)").strip())
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: Optional[bool] = False) -> Optional[str]:
         """
@@ -82,28 +87,32 @@ class FileListSiteUserInfo(SiteParserBase):
         :return: 下页地址
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        size_col = 6
-        seeders_col = 7
+            size_col = 6
+            seeders_col = 7
 
-        page_seeding_size = 0
-        page_seeding_info = []
-        seeding_sizes = html.xpath(f'//table/tr[position()>1]/td[{size_col}]')
-        seeding_seeders = html.xpath(f'//table/tr[position()>1]/td[{seeders_col}]')
-        if seeding_sizes and seeding_seeders:
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = StringUtils.str_int(seeding_seeders[i].xpath("string(.)").strip())
+            page_seeding_size = 0
+            page_seeding_info = []
+            seeding_sizes = html.xpath(f'//table/tr[position()>1]/td[{size_col}]')
+            seeding_seeders = html.xpath(f'//table/tr[position()>1]/td[{seeders_col}]')
+            if seeding_sizes and seeding_seeders:
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = StringUtils.str_int(seeding_seeders[i].xpath("string(.)").strip())
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        self.seeding_info.extend(page_seeding_info)
+            self.seeding_info.extend(page_seeding_info)
 
-        # 是否存在下页数据
-        next_page = None
+            # 是否存在下页数据
+            next_page = None
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 

app/modules/indexer/parser/gazelle.py

@@ -14,46 +14,49 @@ class GazelleSiteUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
+        try:
+            tmps = html.xpath('//a[contains(@href, "user.php?id=")]')
+            if tmps:
+                user_id_match = re.search(r"user.php\?id=(\d+)", tmps[0].attrib['href'])
+                if user_id_match and user_id_match.group().strip():
+                    self.userid = user_id_match.group(1)
+                    self._torrent_seeding_page = f"torrents.php?type=seeding&userid={self.userid}"
+                    self._user_detail_page = f"user.php?id={self.userid}"
+                    self.username = tmps[0].text.strip()
 
-        tmps = html.xpath('//a[contains(@href, "user.php?id=")]')
-        if tmps:
-            user_id_match = re.search(r"user.php\?id=(\d+)", tmps[0].attrib['href'])
-            if user_id_match and user_id_match.group().strip():
-                self.userid = user_id_match.group(1)
-                self._torrent_seeding_page = f"torrents.php?type=seeding&userid={self.userid}"
-                self._user_detail_page = f"user.php?id={self.userid}"
-                self.username = tmps[0].text.strip()
-
-        tmps = html.xpath('//*[@id="header-uploaded-value"]/@data-value')
-        if tmps:
-            self.upload = StringUtils.num_filesize(tmps[0])
-        else:
-            tmps = html.xpath('//li[@id="stats_seeding"]/span/text()')
+            tmps = html.xpath('//*[@id="header-uploaded-value"]/@data-value')
             if tmps:
                 self.upload = StringUtils.num_filesize(tmps[0])
+            else:
+                tmps = html.xpath('//li[@id="stats_seeding"]/span/text()')
+                if tmps:
+                    self.upload = StringUtils.num_filesize(tmps[0])
 
-        tmps = html.xpath('//*[@id="header-downloaded-value"]/@data-value')
-        if tmps:
-            self.download = StringUtils.num_filesize(tmps[0])
-        else:
-            tmps = html.xpath('//li[@id="stats_leeching"]/span/text()')
+            tmps = html.xpath('//*[@id="header-downloaded-value"]/@data-value')
             if tmps:
                 self.download = StringUtils.num_filesize(tmps[0])
+            else:
+                tmps = html.xpath('//li[@id="stats_leeching"]/span/text()')
+                if tmps:
+                    self.download = StringUtils.num_filesize(tmps[0])
 
-        self.ratio = 0.0 if self.download <= 0.0 else round(self.upload / self.download, 3)
+            self.ratio = 0.0 if self.download <= 0.0 else round(self.upload / self.download, 3)
 
-        tmps = html.xpath('//a[contains(@href, "bonus.php")]/@data-tooltip')
-        if tmps:
-            bonus_match = re.search(r"([\d,.]+)", tmps[0])
-            if bonus_match and bonus_match.group(1).strip():
-                self.bonus = StringUtils.str_float(bonus_match.group(1))
-        else:
-            tmps = html.xpath('//a[contains(@href, "bonus.php")]')
+            tmps = html.xpath('//a[contains(@href, "bonus.php")]/@data-tooltip')
             if tmps:
-                bonus_text = tmps[0].xpath("string(.)")
-                bonus_match = re.search(r"([\d,.]+)", bonus_text)
+                bonus_match = re.search(r"([\d,.]+)", tmps[0])
                 if bonus_match and bonus_match.group(1).strip():
                     self.bonus = StringUtils.str_float(bonus_match.group(1))
+            else:
+                tmps = html.xpath('//a[contains(@href, "bonus.php")]')
+                if tmps:
+                    bonus_text = tmps[0].xpath("string(.)")
+                    bonus_match = re.search(r"([\d,.]+)", bonus_text)
+                    if bonus_match and bonus_match.group(1).strip():
+                        self.bonus = StringUtils.str_float(bonus_match.group(1))
+        finally:
+            if html is not None:
+                del html
 
     def _parse_site_page(self, html_text: str):
         pass
@@ -65,27 +68,31 @@ class GazelleSiteUserInfo(SiteParserBase):
         :return:
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        # 用户等级
-        user_levels_text = html.xpath('//*[@id="class-value"]/@data-value')
-        if user_levels_text:
-            self.user_level = user_levels_text[0].strip()
-        else:
-            user_levels_text = html.xpath('//li[contains(text(), "用户等级")]/text()')
+            # 用户等级
+            user_levels_text = html.xpath('//*[@id="class-value"]/@data-value')
             if user_levels_text:
-                self.user_level = user_levels_text[0].split(':')[1].strip()
+                self.user_level = user_levels_text[0].strip()
+            else:
+                user_levels_text = html.xpath('//li[contains(text(), "用户等级")]/text()')
+                if user_levels_text:
+                    self.user_level = user_levels_text[0].split(':')[1].strip()
 
-        # 加入日期
-        join_at_text = html.xpath('//*[@id="join-date-value"]/@data-value')
-        if join_at_text:
-            self.join_at = StringUtils.unify_datetime_str(join_at_text[0].strip())
-        else:
-            join_at_text = html.xpath(
-                '//div[contains(@class, "box_userinfo_stats")]//li[contains(text(), "加入时间")]/span/text()')
+            # 加入日期
+            join_at_text = html.xpath('//*[@id="join-date-value"]/@data-value')
             if join_at_text:
                 self.join_at = StringUtils.unify_datetime_str(join_at_text[0].strip())
+            else:
+                join_at_text = html.xpath(
+                    '//div[contains(@class, "box_userinfo_stats")]//li[contains(text(), "加入时间")]/span/text()')
+                if join_at_text:
+                    self.join_at = StringUtils.unify_datetime_str(join_at_text[0].strip())
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: Optional[bool] = False) -> Optional[str]:
         """
@@ -95,48 +102,52 @@ class GazelleSiteUserInfo(SiteParserBase):
         :return: 下页地址
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        size_col = 3
-        # 搜索size列
-        if html.xpath('//table[contains(@id, "torrent")]//tr[1]/td'):
-            size_col = len(html.xpath('//table[contains(@id, "torrent")]//tr[1]/td')) - 3
-        # 搜索seeders列
-        seeders_col = size_col + 2
+            size_col = 3
+            # 搜索size列
+            if html.xpath('//table[contains(@id, "torrent")]//tr[1]/td'):
+                size_col = len(html.xpath('//table[contains(@id, "torrent")]//tr[1]/td')) - 3
+            # 搜索seeders列
+            seeders_col = size_col + 2
 
-        page_seeding = 0
-        page_seeding_size = 0
-        page_seeding_info = []
-        seeding_sizes = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{size_col}]')
-        seeding_seeders = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{seeders_col}]/text()')
-        if seeding_sizes and seeding_seeders:
-            page_seeding = len(seeding_sizes)
+            page_seeding = 0
+            page_seeding_size = 0
+            page_seeding_info = []
+            seeding_sizes = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{size_col}]')
+            seeding_seeders = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{seeders_col}]/text()')
+            if seeding_sizes and seeding_seeders:
+                page_seeding = len(seeding_sizes)
 
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = int(seeding_seeders[i])
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = int(seeding_seeders[i])
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        if multi_page:
-            self.seeding += page_seeding
-            self.seeding_size += page_seeding_size
-            self.seeding_info.extend(page_seeding_info)
-        else:
-            if not self.seeding:
-                self.seeding = page_seeding
-            if not self.seeding_size:
-                self.seeding_size = page_seeding_size
-            if not self.seeding_info:
-                self.seeding_info = page_seeding_info
+            if multi_page:
+                self.seeding += page_seeding
+                self.seeding_size += page_seeding_size
+                self.seeding_info.extend(page_seeding_info)
+            else:
+                if not self.seeding:
+                    self.seeding = page_seeding
+                if not self.seeding_size:
+                    self.seeding_size = page_seeding_size
+                if not self.seeding_info:
+                    self.seeding_info = page_seeding_info
 
-        # 是否存在下页数据
-        next_page = None
-        next_page_text = html.xpath('//a[contains(.//text(), "Next") or contains(.//text(), "下一页")]/@href')
-        if next_page_text:
-            next_page = next_page_text[-1].strip()
+            # 是否存在下页数据
+            next_page = None
+            next_page_text = html.xpath('//a[contains(.//text(), "Next") or contains(.//text(), "下一页")]/@href')
+            if next_page_text:
+                next_page = next_page_text[-1].strip()
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 

app/modules/indexer/parser/ipt_project.py

@@ -14,67 +14,79 @@ class IptSiteUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
-        tmps = html.xpath('//a[contains(@href, "/u/")]//text()')
-        tmps_id = html.xpath('//a[contains(@href, "/u/")]/@href')
-        if tmps:
-            self.username = str(tmps[-1])
-        if tmps_id:
-            user_id_match = re.search(r"/u/(\d+)", tmps_id[0])
-            if user_id_match and user_id_match.group().strip():
-                self.userid = user_id_match.group(1)
-                self._user_detail_page = f"user.php?u={self.userid}"
-                self._torrent_seeding_page = f"peers?u={self.userid}"
+        try:
+            tmps = html.xpath('//a[contains(@href, "/u/")]//text()')
+            tmps_id = html.xpath('//a[contains(@href, "/u/")]/@href')
+            if tmps:
+                self.username = str(tmps[-1])
+            if tmps_id:
+                user_id_match = re.search(r"/u/(\d+)", tmps_id[0])
+                if user_id_match and user_id_match.group().strip():
+                    self.userid = user_id_match.group(1)
+                    self._user_detail_page = f"user.php?u={self.userid}"
+                    self._torrent_seeding_page = f"peers?u={self.userid}"
 
-        tmps = html.xpath('//div[@class = "stats"]/div/div')
-        if tmps:
-            self.upload = StringUtils.num_filesize(str(tmps[0].xpath('span/text()')[1]).strip())
-            self.download = StringUtils.num_filesize(str(tmps[0].xpath('span/text()')[2]).strip())
-            self.seeding = StringUtils.str_int(tmps[0].xpath('a')[2].xpath('text()')[0])
-            self.leeching = StringUtils.str_int(tmps[0].xpath('a')[2].xpath('text()')[1])
-            self.ratio = StringUtils.str_float(str(tmps[0].xpath('span/text()')[0]).strip().replace('-', '0'))
-            self.bonus = StringUtils.str_float(tmps[0].xpath('a')[3].xpath('text()')[0])
+            tmps = html.xpath('//div[@class = "stats"]/div/div')
+            if tmps:
+                self.upload = StringUtils.num_filesize(str(tmps[0].xpath('span/text()')[1]).strip())
+                self.download = StringUtils.num_filesize(str(tmps[0].xpath('span/text()')[2]).strip())
+                self.seeding = StringUtils.str_int(tmps[0].xpath('a')[2].xpath('text()')[0])
+                self.leeching = StringUtils.str_int(tmps[0].xpath('a')[2].xpath('text()')[1])
+                self.ratio = StringUtils.str_float(str(tmps[0].xpath('span/text()')[0]).strip().replace('-', '0'))
+                self.bonus = StringUtils.str_float(tmps[0].xpath('a')[3].xpath('text()')[0])
+        finally:
+            if html is not None:
+                del html
 
     def _parse_site_page(self, html_text: str):
         pass
 
     def _parse_user_detail_info(self, html_text: str):
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
 
-        user_levels_text = html.xpath('//tr/th[text()="Class"]/following-sibling::td[1]/text()')
-        if user_levels_text:
-            self.user_level = user_levels_text[0].strip()
+            user_levels_text = html.xpath('//tr/th[text()="Class"]/following-sibling::td[1]/text()')
+            if user_levels_text:
+                self.user_level = user_levels_text[0].strip()
 
-        # 加入日期
-        join_at_text = html.xpath('//tr/th[text()="Join date"]/following-sibling::td[1]/text()')
-        if join_at_text:
-            self.join_at = StringUtils.unify_datetime_str(join_at_text[0].split(' (')[0])
+            # 加入日期
+            join_at_text = html.xpath('//tr/th[text()="Join date"]/following-sibling::td[1]/text()')
+            if join_at_text:
+                self.join_at = StringUtils.unify_datetime_str(join_at_text[0].split(' (')[0])
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: bool = False) -> Optional[str]:
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
-        # seeding start
-        seeding_end_pos = 3
-        if html.xpath('//tr/td[text() = "Leechers"]'):
-            seeding_end_pos = len(html.xpath('//tr/td[text() = "Leechers"]/../preceding-sibling::tr')) + 1
-            seeding_end_pos = seeding_end_pos - 3
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
+            # seeding start
+            seeding_end_pos = 3
+            if html.xpath('//tr/td[text() = "Leechers"]'):
+                seeding_end_pos = len(html.xpath('//tr/td[text() = "Leechers"]/../preceding-sibling::tr')) + 1
+                seeding_end_pos = seeding_end_pos - 3
 
-        page_seeding = 0
-        page_seeding_size = 0
-        seeding_torrents = html.xpath('//tr/td[text() = "Seeders"]/../following-sibling::tr/td[position()=6]/text()')
-        if seeding_torrents:
-            page_seeding = seeding_end_pos
-            for per_size in seeding_torrents[:seeding_end_pos]:
-                if '(' in per_size and ')' in per_size:
-                    per_size = per_size.split('(')[-1]
-                    per_size = per_size.split(')')[0]
+            page_seeding = 0
+            page_seeding_size = 0
+            seeding_torrents = html.xpath('//tr/td[text() = "Seeders"]/../following-sibling::tr/td[position()=6]/text()')
+            if seeding_torrents:
+                page_seeding = seeding_end_pos
+                for per_size in seeding_torrents[:seeding_end_pos]:
+                    if '(' in per_size and ')' in per_size:
+                        per_size = per_size.split('(')[-1]
+                        per_size = per_size.split(')')[0]
 
-                page_seeding_size += StringUtils.num_filesize(per_size)
+                    page_seeding_size += StringUtils.num_filesize(per_size)
 
-        self.seeding = page_seeding
-        self.seeding_size = page_seeding_size
+            self.seeding = page_seeding
+            self.seeding_size = page_seeding_size
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_traffic_info(self, html_text: str):
         pass

app/modules/indexer/parser/nexus_audiences.py

@@ -23,12 +23,16 @@ class NexusAudiencesSiteUserInfo(NexusPhpSiteUserInfo):
         if not html_text:
             return
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
-        total_row = html.xpath('//table[@class="table table-bordered"]//tr[td[1][normalize-space()="Total"]]')
-        if not total_row:
-            return
-        seeding_count = total_row[0].xpath('./td[2]/text()')
-        seeding_size = total_row[0].xpath('./td[3]/text()')
-        self.seeding = StringUtils.str_int(seeding_count[0]) if seeding_count else 0
-        self.seeding_size = StringUtils.num_filesize(seeding_size[0].strip()) if seeding_size else 0
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
+            total_row = html.xpath('//table[@class="table table-bordered"]//tr[td[1][normalize-space()="Total"]]')
+            if not total_row:
+                return
+            seeding_count = total_row[0].xpath('./td[2]/text()')
+            seeding_size = total_row[0].xpath('./td[3]/text()')
+            self.seeding = StringUtils.str_int(seeding_count[0]) if seeding_count else 0
+            self.seeding_size = StringUtils.num_filesize(seeding_size[0].strip()) if seeding_size else 0
+        finally:
+            if html is not None:
+                del html

app/modules/indexer/parser/nexus_hhanclub.py

@@ -17,21 +17,25 @@ class NexusHhanclubSiteUserInfo(NexusPhpSiteUserInfo):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
 
-        # 上传、下载、分享率
-        upload_match = re.search(r"[_<>/a-zA-Z-=\"'\s#;]+([\d,.\s]+[KMGTPI]*B)",
-                                 html.xpath('//*[@id="user-info-panel"]/div[2]/div[2]/div[4]/text()')[0])
-        download_match = re.search(r"[_<>/a-zA-Z-=\"'\s#;]+([\d,.\s]+[KMGTPI]*B)",
-                                   html.xpath('//*[@id="user-info-panel"]/div[2]/div[2]/div[5]/text()')[0])
-        ratio_match = re.search(r"分享率][:：_<>/a-zA-Z-=\"'\s#;]+([\d,.\s]+)",
-                                html.xpath('//*[@id="user-info-panel"]/div[2]/div[1]/div[1]/div/text()')[0])
+        try:
+            # 上传、下载、分享率
+            upload_match = re.search(r"[_<>/a-zA-Z-=\"'\s#;]+([\d,.\s]+[KMGTPI]*B)",
+                                     html.xpath('//*[@id="user-info-panel"]/div[2]/div[2]/div[4]/text()')[0])
+            download_match = re.search(r"[_<>/a-zA-Z-=\"'\s#;]+([\d,.\s]+[KMGTPI]*B)",
+                                       html.xpath('//*[@id="user-info-panel"]/div[2]/div[2]/div[5]/text()')[0])
+            ratio_match = re.search(r"分享率][:：_<>/a-zA-Z-=\"'\s#;]+([\d,.\s]+)",
+                                    html.xpath('//*[@id="user-info-panel"]/div[2]/div[1]/div[1]/div/text()')[0])
 
-        # 计算分享率
-        self.upload = StringUtils.num_filesize(upload_match.group(1).strip()) if upload_match else 0
-        self.download = StringUtils.num_filesize(download_match.group(1).strip()) if download_match else 0
-        # 优先使用页面上的分享率
-        calc_ratio = 0.0 if self.download <= 0.0 else round(self.upload / self.download, 3)
-        self.ratio = StringUtils.str_float(ratio_match.group(1)) if (
-                ratio_match and ratio_match.group(1).strip()) else calc_ratio
+            # 计算分享率
+            self.upload = StringUtils.num_filesize(upload_match.group(1).strip()) if upload_match else 0
+            self.download = StringUtils.num_filesize(download_match.group(1).strip()) if download_match else 0
+            # 优先使用页面上的分享率
+            calc_ratio = 0.0 if self.download <= 0.0 else round(self.upload / self.download, 3)
+            self.ratio = StringUtils.str_float(ratio_match.group(1)) if (
+                    ratio_match and ratio_match.group(1).strip()) else calc_ratio
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_detail_info(self, html_text: str):
         """
@@ -42,12 +46,16 @@ class NexusHhanclubSiteUserInfo(NexusPhpSiteUserInfo):
         super()._parse_user_detail_info(html_text)
 
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
-        # 加入时间
-        join_at_text = html.xpath('//*[@id="mainContent"]/div/div[2]/div[4]/div[3]/span[2]/text()[1]')
-        if join_at_text:
-            self.join_at = StringUtils.unify_datetime_str(join_at_text[0].split(' (')[0].strip())
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
+            # 加入时间
+            join_at_text = html.xpath('//*[@id="mainContent"]/div/div[2]/div[4]/div[3]/span[2]/text()[1]')
+            if join_at_text:
+                self.join_at = StringUtils.unify_datetime_str(join_at_text[0].split(' (')[0].strip())
+        finally:
+            if html is not None:
+                del html
 
     def _get_user_level(self, html):
         super()._get_user_level(html)

app/modules/indexer/parser/nexus_php.py

@@ -34,21 +34,25 @@ class NexusPhpSiteUserInfo(SiteParserBase):
         :return:
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
 
-        message_labels = html.xpath('//a[@href="messages.php"]/..')
-        message_labels.extend(html.xpath('//a[contains(@href, "messages.php")]/..'))
-        if message_labels:
-            message_text = message_labels[0].xpath("string(.)")
+            message_labels = html.xpath('//a[@href="messages.php"]/..')
+            message_labels.extend(html.xpath('//a[contains(@href, "messages.php")]/..'))
+            if message_labels:
+                message_text = message_labels[0].xpath("string(.)")
 
-            logger.debug(f"{self._site_name} 消息原始信息 {message_text}")
-            message_unread_match = re.findall(r"[^Date](信息箱\s*|\((?![^)]*:)|你有\xa0)(\d+)", message_text)
+                logger.debug(f"{self._site_name} 消息原始信息 {message_text}")
+                message_unread_match = re.findall(r"[^Date](信息箱\s*|\((?![^)]*:)|你有\xa0)(\d+)", message_text)
 
-            if message_unread_match and len(message_unread_match[-1]) == 2:
-                self.message_unread = StringUtils.str_int(message_unread_match[-1][1])
-            elif message_text.isdigit():
-                self.message_unread = StringUtils.str_int(message_text)
+                if message_unread_match and len(message_unread_match[-1]) == 2:
+                    self.message_unread = StringUtils.str_int(message_unread_match[-1][1])
+                elif message_text.isdigit():
+                    self.message_unread = StringUtils.str_int(message_text)
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_base_info(self, html_text: str):
         """
@@ -61,18 +65,23 @@ class NexusPhpSiteUserInfo(SiteParserBase):
         self._parse_message_unread(html_text)
 
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
 
-        ret = html.xpath(f'//a[contains(@href, "userdetails") and contains(@href, "{self.userid}")]//b//text()')
-        if ret:
-            self.username = str(ret[0])
-            return
-        ret = html.xpath(f'//a[contains(@href, "userdetails") and contains(@href, "{self.userid}")]//text()')
-        if ret:
-            self.username = str(ret[0])
+            ret = html.xpath(f'//a[contains(@href, "userdetails") and contains(@href, "{self.userid}")]//b//text()')
+            if ret:
+                self.username = str(ret[0])
+                return
+            ret = html.xpath(f'//a[contains(@href, "userdetails") and contains(@href, "{self.userid}")]//text()')
+            if ret:
+                self.username = str(ret[0])
+
+            ret = html.xpath('//a[contains(@href, "userdetails")]//strong//text()')
+        finally:
+            if html is not None:
+                del html
 
-        ret = html.xpath('//a[contains(@href, "userdetails")]//strong//text()')
         if ret:
             self.username = str(ret[0])
             return
@@ -98,28 +107,32 @@ class NexusPhpSiteUserInfo(SiteParserBase):
         self.leeching = StringUtils.str_int(leeching_match.group(2)) if leeching_match and leeching_match.group(
             2).strip() else 0
         html = etree.HTML(html_text)
-        has_ucoin, self.bonus = self._parse_ucoin(html)
-        if has_ucoin:
-            return
-        tmps = html.xpath('//a[contains(@href,"mybonus")]/text()') if html else None
-        if tmps:
-            bonus_text = str(tmps[0]).strip()
-            bonus_match = re.search(r"([\d,.]+)", bonus_text)
-            if bonus_match and bonus_match.group(1).strip():
-                self.bonus = StringUtils.str_float(bonus_match.group(1))
-                return
-        bonus_match = re.search(r"mybonus.[\[\]:：<>/a-zA-Z_\-=\"'\s#;.(使用魔力值豆]+\s*([\d,.]+)[<()&\s]", html_text)
         try:
-            if bonus_match and bonus_match.group(1).strip():
-                self.bonus = StringUtils.str_float(bonus_match.group(1))
+            has_ucoin, self.bonus = self._parse_ucoin(html)
+            if has_ucoin:
                 return
-            bonus_match = re.search(r"[魔力值|\]][\[\]:：<>/a-zA-Z_\-=\"'\s#;]+\s*([\d,.]+|\"[\d,.]+\")[<>()&\s]",
-                                    html_text,
-                                    flags=re.S)
-            if bonus_match and bonus_match.group(1).strip():
-                self.bonus = StringUtils.str_float(bonus_match.group(1).strip('"'))
-        except Exception as err:
-            logger.error(f"{self._site_name} 解析魔力值出错, 错误信息: {str(err)}")
+            tmps = html.xpath('//a[contains(@href,"mybonus")]/text()') if html else None
+            if tmps:
+                bonus_text = str(tmps[0]).strip()
+                bonus_match = re.search(r"([\d,.]+)", bonus_text)
+                if bonus_match and bonus_match.group(1).strip():
+                    self.bonus = StringUtils.str_float(bonus_match.group(1))
+                    return
+            bonus_match = re.search(r"mybonus.[\[\]:：<>/a-zA-Z_\-=\"'\s#;.(使用魔力值豆]+\s*([\d,.]+)[<()&\s]", html_text)
+            try:
+                if bonus_match and bonus_match.group(1).strip():
+                    self.bonus = StringUtils.str_float(bonus_match.group(1))
+                    return
+                bonus_match = re.search(r"[魔力值|\]][\[\]:：<>/a-zA-Z_\-=\"'\s#;]+\s*([\d,.]+|\"[\d,.]+\")[<>()&\s]",
+                                        html_text,
+                                        flags=re.S)
+                if bonus_match and bonus_match.group(1).strip():
+                    self.bonus = StringUtils.str_float(bonus_match.group(1).strip('"'))
+            except Exception as err:
+                logger.error(f"{self._site_name} 解析魔力值出错, 错误信息: {str(err)}")
+        finally:
+            if html is not None:
+                del html
 
     @staticmethod
     def _parse_ucoin(html):
@@ -155,72 +168,76 @@ class NexusPhpSiteUserInfo(SiteParserBase):
         :return: 下页地址
         """
         html = etree.HTML(str(html_text).replace(r'\/', '/'))
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        # 首页存在扩展链接，使用扩展链接
-        seeding_url_text = html.xpath('//a[contains(@href,"torrents.php") '
-                                      'and contains(@href,"seeding")]/@href')
-        if multi_page is False and seeding_url_text and seeding_url_text[0].strip():
-            self._torrent_seeding_page = seeding_url_text[0].strip()
-            return self._torrent_seeding_page
+            # 首页存在扩展链接，使用扩展链接
+            seeding_url_text = html.xpath('//a[contains(@href,"torrents.php") '
+                                          'and contains(@href,"seeding")]/@href')
+            if multi_page is False and seeding_url_text and seeding_url_text[0].strip():
+                self._torrent_seeding_page = seeding_url_text[0].strip()
+                return self._torrent_seeding_page
 
-        size_col = 3
-        seeders_col = 4
-        # 搜索size列
-        size_col_xpath = '//tr[position()=1]/' \
-                         'td[(img[@class="size"] and img[@alt="size"])' \
-                         ' or (text() = "大小")' \
-                         ' or (a/img[@class="size" and @alt="size"])]'
-        if html.xpath(size_col_xpath):
-            size_col = len(html.xpath(f'{size_col_xpath}/preceding-sibling::td')) + 1
-        # 搜索seeders列
-        seeders_col_xpath = '//tr[position()=1]/' \
-                            'td[(img[@class="seeders"] and img[@alt="seeders"])' \
-                            ' or (text() = "在做种")' \
-                            ' or (a/img[@class="seeders" and @alt="seeders"])]'
-        if html.xpath(seeders_col_xpath):
-            seeders_col = len(html.xpath(f'{seeders_col_xpath}/preceding-sibling::td')) + 1
+            size_col = 3
+            seeders_col = 4
+            # 搜索size列
+            size_col_xpath = '//tr[position()=1]/' \
+                             'td[(img[@class="size"] and img[@alt="size"])' \
+                             ' or (text() = "大小")' \
+                             ' or (a/img[@class="size" and @alt="size"])]'
+            if html.xpath(size_col_xpath):
+                size_col = len(html.xpath(f'{size_col_xpath}/preceding-sibling::td')) + 1
+            # 搜索seeders列
+            seeders_col_xpath = '//tr[position()=1]/' \
+                                'td[(img[@class="seeders"] and img[@alt="seeders"])' \
+                                ' or (text() = "在做种")' \
+                                ' or (a/img[@class="seeders" and @alt="seeders"])]'
+            if html.xpath(seeders_col_xpath):
+                seeders_col = len(html.xpath(f'{seeders_col_xpath}/preceding-sibling::td')) + 1
 
-        page_seeding = 0
-        page_seeding_size = 0
-        page_seeding_info = []
-        # 如果 table class="torrents"，则增加table[@class="torrents"]
-        table_class = '//table[@class="torrents"]' if html.xpath('//table[@class="torrents"]') else ''
-        seeding_sizes = html.xpath(f'{table_class}//tr[position()>1]/td[{size_col}]')
-        seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]/b/a/text()')
-        if not seeding_seeders:
-            seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]//text()')
-        if seeding_sizes and seeding_seeders:
-            page_seeding = len(seeding_sizes)
+            page_seeding = 0
+            page_seeding_size = 0
+            page_seeding_info = []
+            # 如果 table class="torrents"，则增加table[@class="torrents"]
+            table_class = '//table[@class="torrents"]' if html.xpath('//table[@class="torrents"]') else ''
+            seeding_sizes = html.xpath(f'{table_class}//tr[position()>1]/td[{size_col}]')
+            seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]/b/a/text()')
+            if not seeding_seeders:
+                seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]//text()')
+            if seeding_sizes and seeding_seeders:
+                page_seeding = len(seeding_sizes)
 
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = StringUtils.str_int(seeding_seeders[i])
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = StringUtils.str_int(seeding_seeders[i])
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        self.seeding += page_seeding
-        self.seeding_size += page_seeding_size
-        self.seeding_info.extend(page_seeding_info)
+            self.seeding += page_seeding
+            self.seeding_size += page_seeding_size
+            self.seeding_info.extend(page_seeding_info)
 
-        # 是否存在下页数据
-        next_page = None
-        next_page_text = html.xpath(
-            '//a[contains(.//text(), "下一页") or contains(.//text(), "下一頁") or contains(.//text(), ">")]/@href')
-
-        # 防止识别到详情页
-        while next_page_text:
-            next_page = next_page_text.pop().strip()
-            if not next_page.startswith('details.php'):
-                break
+            # 是否存在下页数据
             next_page = None
+            next_page_text = html.xpath(
+                '//a[contains(.//text(), "下一页") or contains(.//text(), "下一頁") or contains(.//text(), ">")]/@href')
 
-        # fix up page url
-        if next_page:
-            if self.userid not in next_page:
-                next_page = f'{next_page}&userid={self.userid}&type=seeding'
+            # 防止识别到详情页
+            while next_page_text:
+                next_page = next_page_text.pop().strip()
+                if not next_page.startswith('details.php'):
+                    break
+                next_page = None
+
+            # fix up page url
+            if next_page:
+                if self.userid not in next_page:
+                    next_page = f'{next_page}&userid={self.userid}&type=seeding'
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 
@@ -231,57 +248,61 @@ class NexusPhpSiteUserInfo(SiteParserBase):
         :return:
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
 
-        self._get_user_level(html)
+            self._get_user_level(html)
 
-        self._fixup_traffic_info(html)
+            self._fixup_traffic_info(html)
 
-        # 加入日期
-        join_at_text = html.xpath(
-            '//tr/td[text()="加入日期" or text()="注册日期" or *[text()="加入日期"]]/following-sibling::td[1]//text()'
-            '|//div/b[text()="加入日期"]/../text()')
-        if join_at_text:
-            self.join_at = StringUtils.unify_datetime_str(join_at_text[0].split(' (')[0].strip())
+            # 加入日期
+            join_at_text = html.xpath(
+                '//tr/td[text()="加入日期" or text()="注册日期" or *[text()="加入日期"]]/following-sibling::td[1]//text()'
+                '|//div/b[text()="加入日期"]/../text()')
+            if join_at_text:
+                self.join_at = StringUtils.unify_datetime_str(join_at_text[0].split(' (')[0].strip())
 
-        # 做种体积 & 做种数
-        # seeding 页面获取不到的话，此处再获取一次
-        seeding_sizes = html.xpath('//tr/td[text()="当前上传"]/following-sibling::td[1]//'
-                                   'table[tr[1][td[4 and text()="尺寸"]]]//tr[position()>1]/td[4]')
-        seeding_seeders = html.xpath('//tr/td[text()="当前上传"]/following-sibling::td[1]//'
-                                     'table[tr[1][td[5 and text()="做种者"]]]//tr[position()>1]/td[5]//text()')
-        tmp_seeding = len(seeding_sizes)
-        tmp_seeding_size = 0
-        tmp_seeding_info = []
-        for i in range(0, len(seeding_sizes)):
-            size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-            seeders = StringUtils.str_int(seeding_seeders[i])
+            # 做种体积 & 做种数
+            # seeding 页面获取不到的话，此处再获取一次
+            seeding_sizes = html.xpath('//tr/td[text()="当前上传"]/following-sibling::td[1]//'
+                                       'table[tr[1][td[4 and text()="尺寸"]]]//tr[position()>1]/td[4]')
+            seeding_seeders = html.xpath('//tr/td[text()="当前上传"]/following-sibling::td[1]//'
+                                         'table[tr[1][td[5 and text()="做种者"]]]//tr[position()>1]/td[5]//text()')
+            tmp_seeding = len(seeding_sizes)
+            tmp_seeding_size = 0
+            tmp_seeding_info = []
+            for i in range(0, len(seeding_sizes)):
+                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                seeders = StringUtils.str_int(seeding_seeders[i])
 
-            tmp_seeding_size += size
-            tmp_seeding_info.append([seeders, size])
+                tmp_seeding_size += size
+                tmp_seeding_info.append([seeders, size])
 
-        if not self.seeding_size:
-            self.seeding_size = tmp_seeding_size
-        if not self.seeding:
-            self.seeding = tmp_seeding
-        if not self.seeding_info:
-            self.seeding_info = tmp_seeding_info
+            if not self.seeding_size:
+                self.seeding_size = tmp_seeding_size
+            if not self.seeding:
+                self.seeding = tmp_seeding
+            if not self.seeding_info:
+                self.seeding_info = tmp_seeding_info
 
-        seeding_sizes = html.xpath('//tr/td[text()="做种统计"]/following-sibling::td[1]//text()')
-        if seeding_sizes:
-            seeding_match = re.search(r"总做种数:\s+(\d+)", seeding_sizes[0], re.IGNORECASE)
-            seeding_size_match = re.search(r"总做种体积:\s+([\d,.\s]+[KMGTPI]*B)", seeding_sizes[0], re.IGNORECASE)
-            tmp_seeding = StringUtils.str_int(seeding_match.group(1)) if (
-                    seeding_match and seeding_match.group(1)) else 0
-            tmp_seeding_size = StringUtils.num_filesize(
-                seeding_size_match.group(1).strip()) if seeding_size_match else 0
-        if not self.seeding_size:
-            self.seeding_size = tmp_seeding_size
-        if not self.seeding:
-            self.seeding = tmp_seeding
+            seeding_sizes = html.xpath('//tr/td[text()="做种统计"]/following-sibling::td[1]//text()')
+            if seeding_sizes:
+                seeding_match = re.search(r"总做种数:\s+(\d+)", seeding_sizes[0], re.IGNORECASE)
+                seeding_size_match = re.search(r"总做种体积:\s+([\d,.\s]+[KMGTPI]*B)", seeding_sizes[0], re.IGNORECASE)
+                tmp_seeding = StringUtils.str_int(seeding_match.group(1)) if (
+                        seeding_match and seeding_match.group(1)) else 0
+                tmp_seeding_size = StringUtils.num_filesize(
+                    seeding_size_match.group(1).strip()) if seeding_size_match else 0
+            if not self.seeding_size:
+                self.seeding_size = tmp_seeding_size
+            if not self.seeding:
+                self.seeding = tmp_seeding
 
-        self._fixup_torrent_seeding_page(html)
+            self._fixup_torrent_seeding_page(html)
+        finally:
+            if html is not None:
+                del html
 
     def _fixup_torrent_seeding_page(self, html):
         """
@@ -348,43 +369,51 @@ class NexusPhpSiteUserInfo(SiteParserBase):
 
     def _parse_message_unread_links(self, html_text: str, msg_links: list) -> Optional[str]:
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        message_links = html.xpath('//tr[not(./td/img[@alt="Read"])]/td/a[contains(@href, "viewmessage")]/@href')
-        msg_links.extend(message_links)
-        # 是否存在下页数据
-        next_page = None
-        next_page_text = html.xpath('//a[contains(.//text(), "下一页") or contains(.//text(), "下一頁")]/@href')
-        if next_page_text:
-            next_page = next_page_text[-1].strip()
+            message_links = html.xpath('//tr[not(./td/img[@alt="Read"])]/td/a[contains(@href, "viewmessage")]/@href')
+            msg_links.extend(message_links)
+            # 是否存在下页数据
+            next_page = None
+            next_page_text = html.xpath('//a[contains(.//text(), "下一页") or contains(.//text(), "下一頁")]/@href')
+            if next_page_text:
+                next_page = next_page_text[-1].strip()
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 
     def _parse_message_content(self, html_text):
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None, None, None
-        # 标题
-        message_head_text = None
-        message_head = html.xpath('//h1/text()'
-                                  '|//div[@class="layui-card-header"]/span[1]/text()')
-        if message_head:
-            message_head_text = message_head[-1].strip()
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None, None, None
+            # 标题
+            message_head_text = None
+            message_head = html.xpath('//h1/text()'
+                                      '|//div[@class="layui-card-header"]/span[1]/text()')
+            if message_head:
+                message_head_text = message_head[-1].strip()
 
-        # 消息时间
-        message_date_text = None
-        message_date = html.xpath('//h1/following-sibling::table[.//tr/td[@class="colhead"]]//tr[2]/td[2]'
-                                  '|//div[@class="layui-card-header"]/span[2]/span[2]')
-        if message_date:
-            message_date_text = message_date[0].xpath("string(.)").strip()
+            # 消息时间
+            message_date_text = None
+            message_date = html.xpath('//h1/following-sibling::table[.//tr/td[@class="colhead"]]//tr[2]/td[2]'
+                                      '|//div[@class="layui-card-header"]/span[2]/span[2]')
+            if message_date:
+                message_date_text = message_date[0].xpath("string(.)").strip()
 
-        # 消息内容
-        message_content_text = None
-        message_content = html.xpath('//h1/following-sibling::table[.//tr/td[@class="colhead"]]//tr[3]/td'
-                                     '|//div[contains(@class,"layui-card-body")]')
-        if message_content:
-            message_content_text = message_content[0].xpath("string(.)").strip()
+            # 消息内容
+            message_content_text = None
+            message_content = html.xpath('//h1/following-sibling::table[.//tr/td[@class="colhead"]]//tr[3]/td'
+                                         '|//div[contains(@class,"layui-card-body")]')
+            if message_content:
+                message_content_text = message_content[0].xpath("string(.)").strip()
+        finally:
+            if html is not None:
+                del html
 
         return message_head_text, message_date_text, message_content_text
 

app/modules/indexer/parser/nexus_rabbit.py

@@ -114,48 +114,56 @@ class NexusRabbitSiteUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         """只有奶糖余额才需要在 base 中获取，其它均可以在详情页拿到"""
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
-        bonus = html.xpath(
-            '//div[contains(text(), "奶糖余额")]/following-sibling::div[1]/text()'
-        )
-        if bonus:
-            self.bonus = StringUtils.str_float(bonus[0].strip())
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
+            bonus = html.xpath(
+                '//div[contains(text(), "奶糖余额")]/following-sibling::div[1]/text()'
+            )
+            if bonus:
+                self.bonus = StringUtils.str_float(bonus[0].strip())
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_detail_info(self, html_text: str):
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return
-        # 缩小一下查找范围，所有的信息都在这个 div 里
-        user_info = html.xpath('//div[contains(@class, "layui-hares-user-info-right")]')
-        if not user_info:
-            return
-        user_info = user_info[0]
-        # 用户名
-        if username := user_info.xpath(
-            './/span[contains(text(), "用户名")]/a/span/text()'
-        ):
-            self.username = username[0].strip()
-        # 等级
-        if user_level := user_info.xpath('.//span[contains(text(), "等级")]/b/text()'):
-            self.user_level = user_level[0].strip()
-        # 加入日期
-        if join_date := user_info.xpath('.//span[contains(text(), "注册日期")]/text()'):
-            join_date = join_date[0].strip().split("\r")[0].removeprefix("注册日期：")
-            self.join_at = StringUtils.unify_datetime_str(join_date)
-        # 上传量
-        if upload := user_info.xpath('.//span[contains(text(), "上传量")]/text()'):
-            self.upload = StringUtils.num_filesize(
-                upload[0].strip().removeprefix("上传量：")
-            )
-        # 下载量
-        if download := user_info.xpath('.//span[contains(text(), "下载量")]/text()'):
-            self.download = StringUtils.num_filesize(
-                download[0].strip().removeprefix("下载量：")
-            )
-        # 分享率
-        if ratio := user_info.xpath('.//span[contains(text(), "分享率")]/em/text()'):
-            self.ratio = StringUtils.str_float(ratio[0].strip())
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return
+            # 缩小一下查找范围，所有的信息都在这个 div 里
+            user_info = html.xpath('//div[contains(@class, "layui-hares-user-info-right")]')
+            if not user_info:
+                return
+            user_info = user_info[0]
+            # 用户名
+            if username := user_info.xpath(
+                './/span[contains(text(), "用户名")]/a/span/text()'
+            ):
+                self.username = username[0].strip()
+            # 等级
+            if user_level := user_info.xpath('.//span[contains(text(), "等级")]/b/text()'):
+                self.user_level = user_level[0].strip()
+            # 加入日期
+            if join_date := user_info.xpath('.//span[contains(text(), "注册日期")]/text()'):
+                join_date = join_date[0].strip().split("\r")[0].removeprefix("注册日期：")
+                self.join_at = StringUtils.unify_datetime_str(join_date)
+            # 上传量
+            if upload := user_info.xpath('.//span[contains(text(), "上传量")]/text()'):
+                self.upload = StringUtils.num_filesize(
+                    upload[0].strip().removeprefix("上传量：")
+                )
+            # 下载量
+            if download := user_info.xpath('.//span[contains(text(), "下载量")]/text()'):
+                self.download = StringUtils.num_filesize(
+                    download[0].strip().removeprefix("下载量：")
+                )
+            # 分享率
+            if ratio := user_info.xpath('.//span[contains(text(), "分享率")]/em/text()'):
+                self.ratio = StringUtils.str_float(ratio[0].strip())
+        finally:
+            if html is not None:
+                del html
 
     def _parse_message_content(self, html_text):
         """

app/modules/indexer/parser/small_horse.py

@@ -24,9 +24,13 @@ class SmallHorseSiteUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
-        ret = html.xpath('//a[contains(@href, "user.php")]//text()')
-        if ret:
-            self.username = str(ret[0])
+        try:
+            ret = html.xpath('//a[contains(@href, "user.php")]//text()')
+            if ret:
+                self.username = str(ret[0])
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_traffic_info(self, html_text: str):
         """
@@ -36,21 +40,25 @@ class SmallHorseSiteUserInfo(SiteParserBase):
         """
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
-        tmps = html.xpath('//ul[@class = "stats nobullet"]')
-        if tmps:
-            if tmps[1].xpath("li") and tmps[1].xpath("li")[0].xpath("span//text()"):
-                self.join_at = StringUtils.unify_datetime_str(tmps[1].xpath("li")[0].xpath("span//text()")[0])
-            self.upload = StringUtils.num_filesize(str(tmps[1].xpath("li")[2].xpath("text()")[0]).split(":")[1].strip())
-            self.download = StringUtils.num_filesize(
-                str(tmps[1].xpath("li")[3].xpath("text()")[0]).split(":")[1].strip())
-            if tmps[1].xpath("li")[4].xpath("span//text()"):
-                self.ratio = StringUtils.str_float(str(tmps[1].xpath("li")[4].xpath("span//text()")[0]).replace('∞', '0'))
-            else:
-                self.ratio = StringUtils.str_float(str(tmps[1].xpath("li")[5].xpath("text()")[0]).split(":")[1])
-            self.bonus = StringUtils.str_float(str(tmps[1].xpath("li")[5].xpath("text()")[0]).split(":")[1])
-            self.user_level = str(tmps[3].xpath("li")[0].xpath("text()")[0]).split(":")[1].strip()
-            self.leeching = StringUtils.str_int(
-                (tmps[4].xpath("li")[6].xpath("text()")[0]).split(":")[1].replace("[", ""))
+        try:
+            tmps = html.xpath('//ul[@class = "stats nobullet"]')
+            if tmps:
+                if tmps[1].xpath("li") and tmps[1].xpath("li")[0].xpath("span//text()"):
+                    self.join_at = StringUtils.unify_datetime_str(tmps[1].xpath("li")[0].xpath("span//text()")[0])
+                self.upload = StringUtils.num_filesize(str(tmps[1].xpath("li")[2].xpath("text()")[0]).split(":")[1].strip())
+                self.download = StringUtils.num_filesize(
+                    str(tmps[1].xpath("li")[3].xpath("text()")[0]).split(":")[1].strip())
+                if tmps[1].xpath("li")[4].xpath("span//text()"):
+                    self.ratio = StringUtils.str_float(str(tmps[1].xpath("li")[4].xpath("span//text()")[0]).replace('∞', '0'))
+                else:
+                    self.ratio = StringUtils.str_float(str(tmps[1].xpath("li")[5].xpath("text()")[0]).split(":")[1])
+                self.bonus = StringUtils.str_float(str(tmps[1].xpath("li")[5].xpath("text()")[0]).split(":")[1])
+                self.user_level = str(tmps[3].xpath("li")[0].xpath("text()")[0]).split(":")[1].strip()
+                self.leeching = StringUtils.str_int(
+                    (tmps[4].xpath("li")[6].xpath("text()")[0]).split(":")[1].replace("[", ""))
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_detail_info(self, html_text: str):
         pass
@@ -63,39 +71,42 @@ class SmallHorseSiteUserInfo(SiteParserBase):
          :return: 下页地址
          """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        size_col = 6
-        seeders_col = 8
+            size_col = 6
+            seeders_col = 8
 
-        page_seeding = 0
-        page_seeding_size = 0
-        page_seeding_info = []
-        seeding_sizes = html.xpath(f'//table[@id="torrent_table"]//tr[position()>1]/td[{size_col}]')
-        seeding_seeders = html.xpath(f'//table[@id="torrent_table"]//tr[position()>1]/td[{seeders_col}]')
-        if seeding_sizes and seeding_seeders:
-            page_seeding = len(seeding_sizes)
+            page_seeding = 0
+            page_seeding_size = 0
+            page_seeding_info = []
+            seeding_sizes = html.xpath(f'//table[@id="torrent_table"]//tr[position()>1]/td[{size_col}]')
+            seeding_seeders = html.xpath(f'//table[@id="torrent_table"]//tr[position()>1]/td[{seeders_col}]')
+            if seeding_sizes and seeding_seeders:
+                page_seeding = len(seeding_sizes)
 
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = StringUtils.str_int(seeding_seeders[i].xpath("string(.)").strip())
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = StringUtils.str_int(seeding_seeders[i].xpath("string(.)").strip())
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        self.seeding += page_seeding
-        self.seeding_size += page_seeding_size
-        self.seeding_info.extend(page_seeding_info)
-
-        # 是否存在下页数据
-        next_page = None
-        next_pages = html.xpath('//ul[@class="pagination"]/li[contains(@class,"active")]/following-sibling::li')
-        if next_pages and len(next_pages) > 1:
-            page_num = next_pages[0].xpath("string(.)").strip()
-            if page_num.isdigit():
-                next_page = f"{self._torrent_seeding_page}&page={page_num}"
+            self.seeding += page_seeding
+            self.seeding_size += page_seeding_size
+            self.seeding_info.extend(page_seeding_info)
 
+            # 是否存在下页数据
+            next_page = None
+            next_pages = html.xpath('//ul[@class="pagination"]/li[contains(@class,"active")]/following-sibling::li')
+            if next_pages and len(next_pages) > 1:
+                page_num = next_pages[0].xpath("string(.)").strip()
+                if page_num.isdigit():
+                    next_page = f"{self._torrent_seeding_page}&page={page_num}"
+        finally:
+            if html is not None:
+                del html
         return next_page
 
     def _parse_message_unread_links(self, html_text: str, msg_links: list) -> Optional[str]:

app/modules/indexer/parser/torrent_leech.py

@@ -32,29 +32,33 @@ class TorrentLeechSiteUserInfo(SiteParserBase):
         """
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
-        upload_html = html.xpath('//div[contains(@class,"profile-uploaded")]//span/text()')
-        if upload_html:
-            self.upload = StringUtils.num_filesize(upload_html[0])
-        download_html = html.xpath('//div[contains(@class,"profile-downloaded")]//span/text()')
-        if download_html:
-            self.download = StringUtils.num_filesize(download_html[0])
-        ratio_html = html.xpath('//div[contains(@class,"profile-ratio")]//span/text()')
-        if ratio_html:
-            self.ratio = StringUtils.str_float(ratio_html[0].replace('∞', '0'))
+        try:
+            upload_html = html.xpath('//div[contains(@class,"profile-uploaded")]//span/text()')
+            if upload_html:
+                self.upload = StringUtils.num_filesize(upload_html[0])
+            download_html = html.xpath('//div[contains(@class,"profile-downloaded")]//span/text()')
+            if download_html:
+                self.download = StringUtils.num_filesize(download_html[0])
+            ratio_html = html.xpath('//div[contains(@class,"profile-ratio")]//span/text()')
+            if ratio_html:
+                self.ratio = StringUtils.str_float(ratio_html[0].replace('∞', '0'))
 
-        user_level_html = html.xpath('//table[contains(@class, "profileViewTable")]'
-                                     '//tr/td[text()="Class"]/following-sibling::td/text()')
-        if user_level_html:
-            self.user_level = user_level_html[0].strip()
+            user_level_html = html.xpath('//table[contains(@class, "profileViewTable")]'
+                                         '//tr/td[text()="Class"]/following-sibling::td/text()')
+            if user_level_html:
+                self.user_level = user_level_html[0].strip()
 
-        join_at_html = html.xpath('//table[contains(@class, "profileViewTable")]'
-                                  '//tr/td[text()="Registration date"]/following-sibling::td/text()')
-        if join_at_html:
-            self.join_at = StringUtils.unify_datetime_str(join_at_html[0].strip())
+            join_at_html = html.xpath('//table[contains(@class, "profileViewTable")]'
+                                      '//tr/td[text()="Registration date"]/following-sibling::td/text()')
+            if join_at_html:
+                self.join_at = StringUtils.unify_datetime_str(join_at_html[0].strip())
 
-        bonus_html = html.xpath('//span[contains(@class, "total-TL-points")]/text()')
-        if bonus_html:
-            self.bonus = StringUtils.str_float(bonus_html[0].strip())
+            bonus_html = html.xpath('//span[contains(@class, "total-TL-points")]/text()')
+            if bonus_html:
+                self.bonus = StringUtils.str_float(bonus_html[0].strip())
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_detail_info(self, html_text: str):
         pass
@@ -67,33 +71,37 @@ class TorrentLeechSiteUserInfo(SiteParserBase):
         :return: 下页地址
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        size_col = 2
-        seeders_col = 7
+            size_col = 2
+            seeders_col = 7
 
-        page_seeding = 0
-        page_seeding_size = 0
-        page_seeding_info = []
-        seeding_sizes = html.xpath(f'//tbody/tr/td[{size_col}]')
-        seeding_seeders = html.xpath(f'//tbody/tr/td[{seeders_col}]/text()')
-        if seeding_sizes and seeding_seeders:
-            page_seeding = len(seeding_sizes)
+            page_seeding = 0
+            page_seeding_size = 0
+            page_seeding_info = []
+            seeding_sizes = html.xpath(f'//tbody/tr/td[{size_col}]')
+            seeding_seeders = html.xpath(f'//tbody/tr/td[{seeders_col}]/text()')
+            if seeding_sizes and seeding_seeders:
+                page_seeding = len(seeding_sizes)
 
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = StringUtils.str_int(seeding_seeders[i])
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = StringUtils.str_int(seeding_seeders[i])
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        self.seeding += page_seeding
-        self.seeding_size += page_seeding_size
-        self.seeding_info.extend(page_seeding_info)
+            self.seeding += page_seeding
+            self.seeding_size += page_seeding_size
+            self.seeding_info.extend(page_seeding_info)
 
-        # 是否存在下页数据
-        next_page = None
+            # 是否存在下页数据
+            next_page = None
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 

app/modules/indexer/parser/unit3d.py

@@ -14,21 +14,24 @@ class Unit3dSiteUserInfo(SiteParserBase):
     def _parse_user_base_info(self, html_text: str):
         html_text = self._prepare_html_text(html_text)
         html = etree.HTML(html_text)
+        try:
+            tmps = html.xpath('//a[contains(@href, "/users/") and contains(@href, "settings")]/@href')
+            if tmps:
+                user_name_match = re.search(r"/users/(.+)/settings", tmps[0])
+                if user_name_match and user_name_match.group().strip():
+                    self.username = user_name_match.group(1)
+                    self._torrent_seeding_page = f"/users/{self.username}/active?perPage=100&client=&seeding=include"
+                    self._user_detail_page = f"/users/{self.username}"
 
-        tmps = html.xpath('//a[contains(@href, "/users/") and contains(@href, "settings")]/@href')
-        if tmps:
-            user_name_match = re.search(r"/users/(.+)/settings", tmps[0])
-            if user_name_match and user_name_match.group().strip():
-                self.username = user_name_match.group(1)
-                self._torrent_seeding_page = f"/users/{self.username}/active?perPage=100&client=&seeding=include"
-                self._user_detail_page = f"/users/{self.username}"
-
-        tmps = html.xpath('//a[contains(@href, "bonus/earnings")]')
-        if tmps:
-            bonus_text = tmps[0].xpath("string(.)")
-            bonus_match = re.search(r"([\d,.]+)", bonus_text)
-            if bonus_match and bonus_match.group(1).strip():
-                self.bonus = StringUtils.str_float(bonus_match.group(1))
+            tmps = html.xpath('//a[contains(@href, "bonus/earnings")]')
+            if tmps:
+                bonus_text = tmps[0].xpath("string(.)")
+                bonus_match = re.search(r"([\d,.]+)", bonus_text)
+                if bonus_match and bonus_match.group(1).strip():
+                    self.bonus = StringUtils.str_float(bonus_match.group(1))
+        finally:
+            if html is not None:
+                del html
 
     def _parse_site_page(self, html_text: str):
         pass
@@ -40,21 +43,25 @@ class Unit3dSiteUserInfo(SiteParserBase):
         :return:
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        # 用户等级
-        user_levels_text = html.xpath('//div[contains(@class, "content")]//span[contains(@class, "badge-user")]/text()')
-        if user_levels_text:
-            self.user_level = user_levels_text[0].strip()
+            # 用户等级
+            user_levels_text = html.xpath('//div[contains(@class, "content")]//span[contains(@class, "badge-user")]/text()')
+            if user_levels_text:
+                self.user_level = user_levels_text[0].strip()
 
-        # 加入日期
-        join_at_text = html.xpath('//div[contains(@class, "content")]//h4[contains(text(), "注册日期") '
-                                  'or contains(text(), "註冊日期") '
-                                  'or contains(text(), "Registration date")]/text()')
-        if join_at_text:
-            self.join_at = StringUtils.unify_datetime_str(
-                join_at_text[0].replace('注册日期', '').replace('註冊日期', '').replace('Registration date', ''))
+            # 加入日期
+            join_at_text = html.xpath('//div[contains(@class, "content")]//h4[contains(text(), "注册日期") '
+                                      'or contains(text(), "註冊日期") '
+                                      'or contains(text(), "Registration date")]/text()')
+            if join_at_text:
+                self.join_at = StringUtils.unify_datetime_str(
+                    join_at_text[0].replace('注册日期', '').replace('註冊日期', '').replace('Registration date', ''))
+        finally:
+            if html is not None:
+                del html
 
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: Optional[bool] = False) -> Optional[str]:
         """
@@ -64,44 +71,48 @@ class Unit3dSiteUserInfo(SiteParserBase):
         :return: 下页地址
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return None
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return None
 
-        size_col = 9
-        seeders_col = 2
-        # 搜索size列
-        if html.xpath('//thead//th[contains(@class,"size")]'):
-            size_col = len(html.xpath('//thead//th[contains(@class,"size")][1]/preceding-sibling::th')) + 1
-        # 搜索seeders列
-        if html.xpath('//thead//th[contains(@class,"seeders")]'):
-            seeders_col = len(html.xpath('//thead//th[contains(@class,"seeders")]/preceding-sibling::th')) + 1
+            size_col = 9
+            seeders_col = 2
+            # 搜索size列
+            if html.xpath('//thead//th[contains(@class,"size")]'):
+                size_col = len(html.xpath('//thead//th[contains(@class,"size")][1]/preceding-sibling::th')) + 1
+            # 搜索seeders列
+            if html.xpath('//thead//th[contains(@class,"seeders")]'):
+                seeders_col = len(html.xpath('//thead//th[contains(@class,"seeders")]/preceding-sibling::th')) + 1
 
-        page_seeding = 0
-        page_seeding_size = 0
-        page_seeding_info = []
-        seeding_sizes = html.xpath(f'//tr[position()]/td[{size_col}]')
-        seeding_seeders = html.xpath(f'//tr[position()]/td[{seeders_col}]')
-        if seeding_sizes and seeding_seeders:
-            page_seeding = len(seeding_sizes)
+            page_seeding = 0
+            page_seeding_size = 0
+            page_seeding_info = []
+            seeding_sizes = html.xpath(f'//tr[position()]/td[{size_col}]')
+            seeding_seeders = html.xpath(f'//tr[position()]/td[{seeders_col}]')
+            if seeding_sizes and seeding_seeders:
+                page_seeding = len(seeding_sizes)
 
-            for i in range(0, len(seeding_sizes)):
-                size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = StringUtils.str_int(seeding_seeders[i].xpath("string(.)").strip())
+                for i in range(0, len(seeding_sizes)):
+                    size = StringUtils.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                    seeders = StringUtils.str_int(seeding_seeders[i].xpath("string(.)").strip())
 
-                page_seeding_size += size
-                page_seeding_info.append([seeders, size])
+                    page_seeding_size += size
+                    page_seeding_info.append([seeders, size])
 
-        self.seeding += page_seeding
-        self.seeding_size += page_seeding_size
-        self.seeding_info.extend(page_seeding_info)
+            self.seeding += page_seeding
+            self.seeding_size += page_seeding_size
+            self.seeding_info.extend(page_seeding_info)
 
-        # 是否存在下页数据
-        next_page = None
-        next_pages = html.xpath('//ul[@class="pagination"]/li[contains(@class,"active")]/following-sibling::li')
-        if next_pages and len(next_pages) > 1:
-            page_num = next_pages[0].xpath("string(.)").strip()
-            if page_num.isdigit():
-                next_page = f"{self._torrent_seeding_page}&page={page_num}"
+            # 是否存在下页数据
+            next_page = None
+            next_pages = html.xpath('//ul[@class="pagination"]/li[contains(@class,"active")]/following-sibling::li')
+            if next_pages and len(next_pages) > 1:
+                page_num = next_pages[0].xpath("string(.)").strip()
+                if page_num.isdigit():
+                    next_page = f"{self._torrent_seeding_page}&page={page_num}"
+        finally:
+            if html is not None:
+                del html
 
         return next_page
 

app/modules/indexer/spider/__init__.py

@@ -0,0 +1,704 @@
+import datetime
+import re
+import traceback
+from typing import Any, Optional
+from typing import List
+from urllib.parse import quote, urlencode, urlparse, parse_qs
+
+from jinja2 import Template
+from pyquery import PyQuery
+
+from app.core.config import settings
+from app.log import logger
+from app.schemas.types import MediaType
+from app.utils.http import RequestUtils
+from app.utils.string import StringUtils
+
+
+class SiteSpider:
+    """
+    站点爬虫
+    """
+
+    @property
+    def __class__(self):
+        return object
+
+    @property
+    def __dict__(self):
+        return {}
+
+    @property
+    def __dir__(self):
+        raise AttributeError(f"Cannot read protected attribute!")
+
+    def __init__(self,
+                 indexer: dict,
+                 keyword: Optional[str] = None,
+                 mtype: MediaType = None,
+                 cat: Optional[str] = None,
+                 page: Optional[int] = 0,
+                 referer: Optional[str] = None):
+        """
+        设置查询参数
+        :param indexer: 索引器
+        :param keyword: 搜索关键字，如果数组则为批量搜索
+        :param mtype: 媒体类型
+        :param cat: 搜索分类
+        :param page: 页码
+        :param referer: Referer
+        """
+        if not indexer:
+            return
+        self.keyword = keyword
+        self.cat = cat
+        self.mtype = mtype
+        self.indexerid = indexer.get('id')
+        self.indexername = indexer.get('name')
+        self.search = indexer.get('search')
+        self.batch = indexer.get('batch')
+        self.browse = indexer.get('browse')
+        self.category = indexer.get('category')
+        self.list = indexer.get('torrents').get('list', {})
+        self.fields = indexer.get('torrents').get('fields')
+        if not keyword and self.browse:
+            self.list = self.browse.get('list') or self.list
+            self.fields = self.browse.get('fields') or self.fields
+        self.domain = indexer.get('domain')
+        self.result_num = int(indexer.get('result_num') or 100)
+        self._timeout = int(indexer.get('timeout') or 15)
+        self.page = page
+        if self.domain and not str(self.domain).endswith("/"):
+            self.domain = self.domain + "/"
+        self.ua = indexer.get('ua') or settings.USER_AGENT
+        self.proxies = settings.PROXY if indexer.get('proxy') else None
+        self.proxy_server = settings.PROXY_SERVER if indexer.get('proxy') else None
+        self.cookie = indexer.get('cookie')
+        self.referer = referer
+        # 初始化属性
+        self.is_error = False
+        self.torrents_info = {}
+        self.torrents_info_array = []
+
+    def get_torrents(self) -> List[dict]:
+        """
+        开始请求
+        """
+        if not self.search or not self.domain:
+            return []
+
+        # 种子搜索相对路径
+        paths = self.search.get('paths', [])
+        torrentspath = ""
+        if len(paths) == 1:
+            torrentspath = paths[0].get('path', '')
+        else:
+            for path in paths:
+                if path.get("type") == "all" and not self.mtype:
+                    torrentspath = path.get('path')
+                    break
+                elif path.get("type") == "movie" and self.mtype == MediaType.MOVIE:
+                    torrentspath = path.get('path')
+                    break
+                elif path.get("type") == "tv" and self.mtype == MediaType.TV:
+                    torrentspath = path.get('path')
+                    break
+
+        # 精确搜索
+        if self.keyword:
+            if isinstance(self.keyword, list):
+                # 批量查询
+                if self.batch:
+                    delimiter = self.batch.get('delimiter') or ' '
+                    space_replace = self.batch.get('space_replace') or ' '
+                    search_word = delimiter.join([str(k).replace(' ',
+                                                                 space_replace) for k in self.keyword])
+                else:
+                    search_word = " ".join(self.keyword)
+                # 查询模式：或
+                search_mode = "1"
+            else:
+                # 单个查询
+                search_word = self.keyword
+                # 查询模式与
+                search_mode = "0"
+
+            # 搜索URL
+            indexer_params = self.search.get("params", {}).copy()
+            if indexer_params:
+                search_area = indexer_params.get('search_area')
+                # search_area非0表示支持imdbid搜索
+                if (search_area and
+                        (not self.keyword or not self.keyword.startswith('tt'))):
+                    # 支持imdbid搜索，但关键字不是imdbid时，不启用imdbid搜索
+                    indexer_params.pop('search_area')
+                # 变量字典
+                inputs_dict = {
+                    "keyword": search_word
+                }
+                # 查询参数，默认查询标题
+                params = {
+                    "search_mode": search_mode,
+                    "search_area": 0,
+                    "page": self.page or 0,
+                    "notnewword": 1
+                }
+                # 额外参数
+                for key, value in indexer_params.items():
+                    params.update({
+                        "%s" % key: str(value).format(**inputs_dict)
+                    })
+                # 分类条件
+                if self.category:
+                    if self.mtype == MediaType.TV:
+                        cats = self.category.get("tv") or []
+                    elif self.mtype == MediaType.MOVIE:
+                        cats = self.category.get("movie") or []
+                    else:
+                        cats = (self.category.get("movie") or []) + (self.category.get("tv") or [])
+                    for cat in cats:
+                        if self.cat and str(cat.get("id")) not in self.cat:
+                            continue
+                        if self.category.get("field"):
+                            value = params.get(self.category.get("field"), "")
+                            params.update({
+                                "%s" % self.category.get("field"): value + self.category.get("delimiter",
+                                                                                             ' ') + cat.get("id")
+                            })
+                        else:
+                            params.update({
+                                "cat%s" % cat.get("id"): 1
+                            })
+                searchurl = self.domain + torrentspath + "?" + urlencode(params)
+            else:
+                # 变量字典
+                inputs_dict = {
+                    "keyword": quote(search_word),
+                    "page": self.page or 0
+                }
+                # 无额外参数
+                searchurl = self.domain + str(torrentspath).format(**inputs_dict)
+
+        # 列表浏览
+        else:
+            # 变量字典
+            inputs_dict = {
+                "page": self.page or 0,
+                "keyword": ""
+            }
+            # 有单独浏览路径
+            if self.browse:
+                torrentspath = self.browse.get("path")
+                if self.browse.get("start"):
+                    start_page = int(self.browse.get("start")) + int(self.page or 0)
+                    inputs_dict.update({
+                        "page": start_page
+                    })
+            elif self.page:
+                torrentspath = torrentspath + f"?page={self.page}"
+            # 搜索Url
+            searchurl = self.domain + str(torrentspath).format(**inputs_dict)
+
+        logger.info(f"开始请求：{searchurl}")
+
+        # requests请求
+        ret = RequestUtils(
+            ua=self.ua,
+            cookies=self.cookie,
+            timeout=self._timeout,
+            referer=self.referer,
+            proxies=self.proxies
+        ).get_res(searchurl, allow_redirects=True)
+        # 解析返回
+        return self.parse(
+            RequestUtils.get_decoded_html_content(
+                ret,
+                performance_mode=settings.ENCODING_DETECTION_PERFORMANCE_MODE,
+                confidence_threshold=settings.ENCODING_DETECTION_MIN_CONFIDENCE
+            )
+        )
+
+    def __get_title(self, torrent: Any):
+        # title default text
+        if 'title' not in self.fields:
+            return
+        selector = self.fields.get('title', {})
+        if 'selector' in selector:
+            self.torrents_info['title'] = self._safe_query(torrent, selector)
+        elif 'text' in selector:
+            render_dict = {}
+            if "title_default" in self.fields:
+                title_default_selector = self.fields.get('title_default', {})
+                title_default = self._safe_query(torrent, title_default_selector)
+                render_dict.update({'title_default': title_default})
+            if "title_optional" in self.fields:
+                title_optional_selector = self.fields.get('title_optional', {})
+                title_optional = self._safe_query(torrent, title_optional_selector)
+                render_dict.update({'title_optional': title_optional})
+            self.torrents_info['title'] = Template(selector.get('text')).render(fields=render_dict)
+        self.torrents_info['title'] = self.__filter_text(self.torrents_info.get('title'),
+                                                         selector.get('filters'))
+
+    def __get_description(self, torrent: Any):
+        # description text
+        if 'description' not in self.fields:
+            return
+        selector = self.fields.get('description', {})
+        if "selector" in selector or "selectors" in selector:
+            # 对于selectors情况，需要特殊处理selector_config
+            desc_selector = selector.copy()
+            if "selectors" in selector and "selector" not in selector:
+                desc_selector["selector"] = selector.get("selectors", "")
+            self.torrents_info['description'] = self._safe_query(torrent, desc_selector)
+        elif "text" in selector:
+            render_dict = {}
+            if "tags" in self.fields:
+                tags_selector = self.fields.get('tags', {})
+                tag = self._safe_query(torrent, tags_selector)
+                render_dict.update({'tags': tag})
+            if "subject" in self.fields:
+                subject_selector = self.fields.get('subject', {})
+                subject = self._safe_query(torrent, subject_selector)
+                render_dict.update({'subject': subject})
+            if "description_free_forever" in self.fields:
+                description_free_forever_selector = self.fields.get("description_free_forever", {})
+                description_free_forever = self._safe_query(torrent, description_free_forever_selector)
+                render_dict.update({"description_free_forever": description_free_forever})
+            if "description_normal" in self.fields:
+                description_normal_selector = self.fields.get("description_normal", {})
+                description_normal = self._safe_query(torrent, description_normal_selector)
+                render_dict.update({"description_normal": description_normal})
+            self.torrents_info['description'] = Template(selector.get('text')).render(fields=render_dict)
+        self.torrents_info['description'] = self.__filter_text(self.torrents_info.get('description'),
+                                                               selector.get('filters'))
+
+    def __get_detail(self, torrent: Any):
+        # details page text
+        if 'details' not in self.fields:
+            return
+        selector = self.fields.get('details', {})
+        item = self._safe_query(torrent, selector)
+        detail_link = self.__filter_text(item, selector.get('filters'))
+        if detail_link:
+            if not detail_link.startswith("http"):
+                if detail_link.startswith("//"):
+                    self.torrents_info['page_url'] = self.domain.split(":")[0] + ":" + detail_link
+                elif detail_link.startswith("/"):
+                    self.torrents_info['page_url'] = self.domain + detail_link[1:]
+                else:
+                    self.torrents_info['page_url'] = self.domain + detail_link
+            else:
+                self.torrents_info['page_url'] = detail_link
+
+    def __get_download(self, torrent: Any):
+        # download link text
+        if 'download' not in self.fields:
+            return
+        selector = self.fields.get('download', {})
+        item = self._safe_query(torrent, selector)
+        download_link = self.__filter_text(item, selector.get('filters'))
+        if download_link:
+            if not download_link.startswith("http") \
+                    and not download_link.startswith("magnet"):
+                _scheme, _domain = StringUtils.get_url_netloc(self.domain)
+                if _domain in download_link:
+                    if download_link.startswith("/"):
+                        self.torrents_info['enclosure'] = f"{_scheme}:{download_link}"
+                    else:
+                        self.torrents_info['enclosure'] = f"{_scheme}://{download_link}"
+                else:
+                    if download_link.startswith("/"):
+                        self.torrents_info['enclosure'] = f"{self.domain}{download_link[1:]}"
+                    else:
+                        self.torrents_info['enclosure'] = f"{self.domain}{download_link}"
+            else:
+                self.torrents_info['enclosure'] = download_link
+
+    def __get_imdbid(self, torrent: Any):
+        # imdbid
+        if "imdbid" not in self.fields:
+            return
+        selector = self.fields.get('imdbid', {})
+        item = self._safe_query(torrent, selector)
+        self.torrents_info['imdbid'] = self.__filter_text(item, selector.get('filters'))
+
+    def __get_size(self, torrent: Any):
+        # torrent size int
+        if 'size' not in self.fields:
+            return
+        selector = self.fields.get('size', {})
+        item = self._safe_query(torrent, selector)
+        if item:
+            size_val = item.replace("\n", "").strip()
+            size_val = self.__filter_text(size_val,
+                                          selector.get('filters'))
+            self.torrents_info['size'] = StringUtils.num_filesize(size_val)
+        else:
+            self.torrents_info['size'] = 0
+
+    def __get_leechers(self, torrent: Any):
+        # torrent leechers int
+        if 'leechers' not in self.fields:
+            return
+        selector = self.fields.get('leechers', {})
+        item = self._safe_query(torrent, selector)
+        if item:
+            peers_val = item.split("/")[0]
+            peers_val = peers_val.replace(",", "")
+            peers_val = self.__filter_text(peers_val, selector.get('filters'))
+            self.torrents_info['peers'] = int(peers_val) if peers_val and peers_val.isdigit() else 0
+        else:
+            self.torrents_info['peers'] = 0
+
+    def __get_seeders(self, torrent: Any):
+        # torrent seeders int
+        if 'seeders' not in self.fields:
+            return
+        selector = self.fields.get('seeders', {})
+        item = self._safe_query(torrent, selector)
+        if item:
+            seeders_val = item.split("/")[0]
+            seeders_val = seeders_val.replace(",", "")
+            seeders_val = self.__filter_text(seeders_val, selector.get('filters'))
+            self.torrents_info['seeders'] = int(seeders_val) if seeders_val and seeders_val.isdigit() else 0
+        else:
+            self.torrents_info['seeders'] = 0
+
+    def __get_grabs(self, torrent: Any):
+        # torrent grabs int
+        if 'grabs' not in self.fields:
+            return
+        selector = self.fields.get('grabs', {})
+        item = self._safe_query(torrent, selector)
+        if item:
+            grabs_val = item.split("/")[0]
+            grabs_val = grabs_val.replace(",", "")
+            grabs_val = self.__filter_text(grabs_val, selector.get('filters'))
+            self.torrents_info['grabs'] = int(grabs_val) if grabs_val and grabs_val.isdigit() else 0
+        else:
+            self.torrents_info['grabs'] = 0
+
+    def __get_pubdate(self, torrent: Any):
+        # torrent pubdate yyyy-mm-dd hh:mm:ss
+        if 'date_added' not in self.fields:
+            return
+        selector = self.fields.get('date_added', {})
+        pubdate_str = self._safe_query(torrent, selector)
+        if pubdate_str:
+            pubdate_str = pubdate_str.replace('\n', ' ').strip()
+        self.torrents_info['pubdate'] = self.__filter_text(pubdate_str, selector.get('filters'))
+
+    def __get_date_elapsed(self, torrent: Any):
+        # torrent date elapsed text
+        if 'date_elapsed' not in self.fields:
+            return
+        selector = self.fields.get('date_elapsed', {})
+        date_elapsed = self._safe_query(torrent, selector)
+        self.torrents_info['date_elapsed'] = self.__filter_text(date_elapsed, selector.get('filters'))
+
+    def __get_downloadvolumefactor(self, torrent: Any):
+        # downloadvolumefactor int
+        selector = self.fields.get('downloadvolumefactor', {})
+        if not selector:
+            return
+        self.torrents_info['downloadvolumefactor'] = 1
+        if 'case' in selector:
+            for downloadvolumefactorselector in list(selector.get('case', {}).keys()):
+                downloadvolumefactor = torrent(downloadvolumefactorselector)
+                try:
+                    if len(downloadvolumefactor) > 0:
+                        self.torrents_info['downloadvolumefactor'] = selector.get('case', {}).get(
+                            downloadvolumefactorselector)
+                        break
+                finally:
+                    downloadvolumefactor.clear()
+                    del downloadvolumefactor
+        elif "selector" in selector:
+            item = self._safe_query(torrent, selector)
+            if item:
+                downloadvolumefactor = re.search(r'(\d+\.?\d*)', item)
+                if downloadvolumefactor:
+                    self.torrents_info['downloadvolumefactor'] = int(downloadvolumefactor.group(1))
+
+    def __get_uploadvolumefactor(self, torrent: Any):
+        # uploadvolumefactor int
+        selector = self.fields.get('uploadvolumefactor', {})
+        if not selector:
+            return
+        self.torrents_info['uploadvolumefactor'] = 1
+        if 'case' in selector:
+            for uploadvolumefactorselector in list(selector.get('case', {}).keys()):
+                uploadvolumefactor = torrent(uploadvolumefactorselector)
+                try:
+                    if len(uploadvolumefactor) > 0:
+                        self.torrents_info['uploadvolumefactor'] = selector.get('case', {}).get(
+                            uploadvolumefactorselector)
+                        break
+                finally:
+                    uploadvolumefactor.clear()
+                    del uploadvolumefactor
+        elif "selector" in selector:
+            item = self._safe_query(torrent, selector)
+            if item:
+                uploadvolumefactor = re.search(r'(\d+\.?\d*)', item)
+                if uploadvolumefactor:
+                    self.torrents_info['uploadvolumefactor'] = int(uploadvolumefactor.group(1))
+
+    def __get_labels(self, torrent: Any):
+        # labels ['label1', 'label2']
+        if 'labels' not in self.fields:
+            return
+        selector = self.fields.get('labels', {})
+        if not selector.get('selector'):
+            self.torrents_info['labels'] = []
+            return
+
+        # labels需要特殊处理，因为它返回的是列表
+        labels = torrent(selector.get("selector", "")).clone()
+        try:
+            self.__remove(labels, selector)
+            items = self.__attribute_or_text(labels, selector)
+            if items:
+                self.torrents_info['labels'] = [item for item in items if item]
+            else:
+                self.torrents_info['labels'] = []
+        finally:
+            labels.clear()
+            del labels
+
+    def __get_free_date(self, torrent: Any):
+        # free date yyyy-mm-dd hh:mm:ss
+        if 'freedate' not in self.fields:
+            return
+        selector = self.fields.get('freedate', {})
+        freedate = self._safe_query(torrent, selector)
+        self.torrents_info['freedate'] = self.__filter_text(freedate, selector.get('filters'))
+
+    def __get_hit_and_run(self, torrent: Any):
+        # hitandrun True/False
+        if 'hr' not in self.fields:
+            return
+        selector = self.fields.get('hr', {})
+        hit_and_run = torrent(selector.get('selector', ''))
+        try:
+            if hit_and_run:
+                self.torrents_info['hit_and_run'] = True
+            else:
+                self.torrents_info['hit_and_run'] = False
+        finally:
+            hit_and_run.clear()
+            del hit_and_run
+
+    def __get_category(self, torrent: Any):
+        # category 电影/电视剧
+        if 'category' not in self.fields:
+            return
+        selector = self.fields.get('category', {})
+        category_value = self._safe_query(torrent, selector)
+        category_value = self.__filter_text(category_value, selector.get('filters'))
+        if category_value and self.category:
+            tv_cats = [str(cat.get("id")) for cat in self.category.get("tv") or []]
+            movie_cats = [str(cat.get("id")) for cat in self.category.get("movie") or []]
+            if category_value in tv_cats \
+                    and category_value not in movie_cats:
+                self.torrents_info['category'] = MediaType.TV.value
+            elif category_value in movie_cats:
+                self.torrents_info['category'] = MediaType.MOVIE.value
+            else:
+                self.torrents_info['category'] = MediaType.UNKNOWN.value
+        else:
+            self.torrents_info['category'] = MediaType.UNKNOWN.value
+
+    def _safe_query(self, torrent: Any, selector_config: Optional[dict]) -> Optional[str]:
+        """
+        安全地执行PyQuery查询并自动清理资源
+        :param torrent: PyQuery对象
+        :param selector_config: 选择器配置
+        :return: 处理后的结果
+        """
+        if not selector_config or not selector_config.get('selector'):
+            return None
+
+        query_obj = torrent(selector_config.get('selector', '')).clone()
+        try:
+            self.__remove(query_obj, selector_config)
+            items = self.__attribute_or_text(query_obj, selector_config)
+            return self.__index(items, selector_config)
+        finally:
+            query_obj.clear()
+            del query_obj
+
+    def get_info(self, torrent: Any) -> dict:
+        """
+        解析单条种子数据
+        """
+        # 每次调用时重新初始化，避免数据累积
+        self.torrents_info = {}
+        try:
+            # 标题
+            self.__get_title(torrent)
+            # 描述
+            self.__get_description(torrent)
+            # 详情页面
+            self.__get_detail(torrent)
+            # 下载链接
+            self.__get_download(torrent)
+            # 完成数
+            self.__get_grabs(torrent)
+            # 下载数
+            self.__get_leechers(torrent)
+            # 做种数
+            self.__get_seeders(torrent)
+            # 大小
+            self.__get_size(torrent)
+            # IMDBID
+            self.__get_imdbid(torrent)
+            # 下载系数
+            self.__get_downloadvolumefactor(torrent)
+            # 上传系数
+            self.__get_uploadvolumefactor(torrent)
+            # 发布时间
+            self.__get_pubdate(torrent)
+            # 已发布时间
+            self.__get_date_elapsed(torrent)
+            # 免费载止时间
+            self.__get_free_date(torrent)
+            # 标签
+            self.__get_labels(torrent)
+            # HR
+            self.__get_hit_and_run(torrent)
+            # 分类
+            self.__get_category(torrent)
+            # 返回当前种子信息的副本，而不是引用
+            return self.torrents_info.copy() if self.torrents_info else {}
+        except Exception as err:
+            logger.error("%s 搜索出现错误：%s" % (self.indexername, str(err)))
+            return {}
+        finally:
+            self.torrents_info.clear()
+
+    @staticmethod
+    def __filter_text(text: Optional[str], filters: Optional[List[dict]]) -> str:
+        """
+        对文件进行处理
+        """
+        if not text or not filters or not isinstance(filters, list):
+            return text
+        if not isinstance(text, str):
+            text = str(text)
+        for filter_item in filters:
+            if not text:
+                break
+            method_name = filter_item.get("name")
+            try:
+                args = filter_item.get("args")
+                if method_name == "re_search" and isinstance(args, list):
+                    rematch = re.search(r"%s" % args[0], text)
+                    if rematch:
+                        text = rematch.group(args[-1])
+                elif method_name == "split" and isinstance(args, list):
+                    text = text.split(r"%s" % args[0])[args[-1]]
+                elif method_name == "replace" and isinstance(args, list):
+                    text = text.replace(r"%s" % args[0], r"%s" % args[-1])
+                elif method_name == "dateparse" and isinstance(args, str):
+                    text = text.replace("\n", " ").strip()
+                    text = datetime.datetime.strptime(text, r"%s" % args)
+                elif method_name == "strip":
+                    text = text.strip()
+                elif method_name == "appendleft":
+                    text = f"{args}{text}"
+                elif method_name == "querystring":
+                    parsed_url = urlparse(str(text))
+                    query_params = parse_qs(parsed_url.query)
+                    param_value = query_params.get(args)
+                    text = param_value[0] if param_value else ''
+            except Exception as err:
+                logger.debug(f'过滤器 {method_name} 处理失败：{str(err)} - {traceback.format_exc()}')
+        return text.strip()
+
+    @staticmethod
+    def __remove(item: Any, selector: Optional[dict]):
+        """
+        移除元素
+        """
+        if selector and "remove" in selector:
+            removelist = selector.get('remove', '').split(', ')
+            for v in removelist:
+                item.remove(v)
+
+    @staticmethod
+    def __attribute_or_text(item: Any, selector: Optional[dict]) -> list:
+        if not selector:
+            return item
+        if not item:
+            return []
+        if 'attribute' in selector:
+            items = [i.attr(selector.get('attribute')) for i in item.items() if i]
+        else:
+            items = [i.text() for i in item.items() if i]
+        return items
+
+    @staticmethod
+    def __index(items: Optional[list], selector: Optional[dict]) -> Optional[str]:
+        if not items:
+            return None
+        if selector:
+            if "contents" in selector \
+                    and len(items) > int(selector.get("contents")):
+                item = items[0].split("\n")[selector.get("contents")]
+            elif "index" in selector \
+                    and len(items) > int(selector.get("index")):
+                item = items[int(selector.get("index"))]
+            else:
+                item = items[0]
+        else:
+            item = items[0]
+        return item
+
+    def parse(self, html_text: str) -> List[dict]:
+        """
+        解析整个页面
+        """
+        if not html_text:
+            self.is_error = True
+            return []
+
+        # 清空旧结果
+        self.torrents_info_array = []
+        html_doc = None
+        try:
+            # 解析站点文本对象
+            html_doc = PyQuery(html_text)
+            # 种子筛选器
+            torrents_selector = self.list.get('selector', '')
+            # 遍历种子html列表
+            for i, torn in enumerate(html_doc(torrents_selector)):
+                if i >= int(self.result_num):
+                    break
+                # 创建临时PyQuery对象进行解析
+                torrent_query = PyQuery(torn)
+                try:
+                    # 直接获取种子信息，避免深拷贝
+                    torrent_info = self.get_info(torrent_query)
+                    if torrent_info:
+                        # 浅拷贝即可，减少内存使用
+                        self.torrents_info_array.append(torrent_info)
+                finally:
+                    # 显式删除临时PyQuery对象
+                    torrent_query.clear()
+                    del torrent_query
+            # 返回数组的副本，防止被后续清理操作影响
+            return self.torrents_info_array.copy()
+        except Exception as err:
+            self.is_error = True
+            logger.warn(f"错误：{self.indexername} {str(err)}")
+            return []
+        finally:
+            # 清理种子缓存
+            self.torrents_info_array.clear()
+            # 清理HTML文档对象
+            if html_doc is not None:
+                html_doc.clear()
+                del html_doc
+            # 清理html_text引用
+            del html_text

app/modules/subtitle/__init__.py

@@ -104,20 +104,24 @@ class SubtitleModule(_ModuleBase):
                 logger.warn(f"读取页面代码失败：{torrent.page_url}")
                 return
             html = etree.HTML(res.text)
-            sublink_list = []
-            for xpath in self._SITE_SUBTITLE_XPATH:
-                sublinks = html.xpath(xpath)
-                if sublinks:
-                    for sublink in sublinks:
-                        if not sublink:
-                            continue
-                        if not sublink.startswith("http"):
-                            base_url = StringUtils.get_base_url(torrent.page_url)
-                            if sublink.startswith("/"):
-                                sublink = "%s%s" % (base_url, sublink)
-                            else:
-                                sublink = "%s/%s" % (base_url, sublink)
-                        sublink_list.append(sublink)
+            try:
+                sublink_list = []
+                for xpath in self._SITE_SUBTITLE_XPATH:
+                    sublinks = html.xpath(xpath)
+                    if sublinks:
+                        for sublink in sublinks:
+                            if not sublink:
+                                continue
+                            if not sublink.startswith("http"):
+                                base_url = StringUtils.get_base_url(torrent.page_url)
+                                if sublink.startswith("/"):
+                                    sublink = "%s%s" % (base_url, sublink)
+                                else:
+                                    sublink = "%s/%s" % (base_url, sublink)
+                            sublink_list.append(sublink)
+            finally:
+                if html is not None:
+                    del html
             # 下载所有字幕文件
             for sublink in sublink_list:
                 logger.info(f"找到字幕下载链接：{sublink}，开始下载...")

app/modules/themoviedb/tmdbapi.py

@@ -563,6 +563,9 @@ class TmdbApi:
         except Exception as err:
             logger.error(f"从TheDbMovie网站查询出错：{str(err)}")
             return {}
+        finally:
+            if html is not None:
+                del html
         return {}
 
     def get_info(self,

app/modules/transmission/transmission.py

@@ -25,7 +25,7 @@ class Transmission:
         若不设置参数，则创建配置文件设置的下载器
         """
         if host and port:
-            self._protocol, self._host, self._port = kwargs.get("protocol", self._protocol), host, port
+            self._protocol, self._host, self._port = kwargs.get("protocol", "http"), host, port
         elif host:
             result = UrlUtils.parse_url_params(url=host)
             if result:

app/modules/trimemedia/api.py

@@ -111,7 +111,7 @@ class Api:
         "_api_path",
         "_request_utils",
         "_version",
-        "_session"
+        "_session",
     )
 
     @property
@@ -287,6 +287,18 @@ class Api:
                 return True
         return False
 
+    def task_running(self):
+        """
+        当前正在运行的任务
+        """
+        if (
+            res := self.__request_api("/task/running")
+        ) and res.success:
+            if res.data:
+                # TODO 具体正在运行的任务
+                return True
+        return False
+
     def __build_item(self, info: dict) -> Item:
         """
         构造媒体Item

app/modules/trimemedia/trimemedia.py

@@ -111,6 +111,8 @@ class TrimeMedia:
         if self._userinfo is None:
             return False
         logger.debug(f"{self._username} 成功登录飞牛影视")
+        # 刷新媒体库列表
+        self.get_librarys()
         return True
 
     def disconnect(self):
@@ -311,6 +313,8 @@ class TrimeMedia:
             logger.error("飞牛仅支持管理员账号刷新媒体库")
             return False
 
+        # 必须调用 否则容易误报 -14 Task duplicate
+        self._api.task_running()
         logger.info("刷新所有媒体库")
         return self._api.mdb_scanall()
 
@@ -337,6 +341,8 @@ class TrimeMedia:
             # 媒体库去重
             libraries.add(lib.guid)
 
+        # 必须调用 否则容易误报 -14 Task duplicate
+        self._api.task_running()
         for lib_guid in libraries:
             # 逐个刷新
             lib = self._libraries[lib_guid]

app/utils/site.py

@@ -13,27 +13,31 @@ class SiteUtils:
         :return:
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
+        try:
+            if not StringUtils.is_valid_html_element(html):
+                return False
+            # 存在明显的密码输入框，说明未登录
+            if html.xpath("//input[@type='password']"):
+                return False
+            # 是否存在登出和用户面板等链接
+            xpaths = [
+                '//a[contains(@href, "logout")'
+                ' or contains(@data-url, "logout")'
+                ' or contains(@href, "mybonus") '
+                ' or contains(@onclick, "logout")'
+                ' or contains(@href, "usercp")'
+                ' or contains(@lay-on, "logout")]',
+                '//form[contains(@action, "logout")]',
+                '//div[@class="user-info-side"]',
+                '//a[@id="myitem"]'
+            ]
+            for xpath in xpaths:
+                if html.xpath(xpath):
+                    return True
             return False
-        # 存在明显的密码输入框，说明未登录
-        if html.xpath("//input[@type='password']"):
-            return False
-        # 是否存在登出和用户面板等链接
-        xpaths = [
-            '//a[contains(@href, "logout")'
-            ' or contains(@data-url, "logout")'
-            ' or contains(@href, "mybonus") '
-            ' or contains(@onclick, "logout")'
-            ' or contains(@href, "usercp")'
-            ' or contains(@lay-on, "logout")]',
-            '//form[contains(@action, "logout")]',
-            '//div[@class="user-info-side"]',
-            '//a[@id="myitem"]'
-        ]
-        for xpath in xpaths:
-            if html.xpath(xpath):
-                return True
-        return False
+        finally:
+            if html is not None:
+                del html
 
     @classmethod
     def is_checkin(cls, html_text: str) -> bool:
@@ -42,24 +46,27 @@ class SiteUtils:
         :return True已签到 False未签到
         """
         html = etree.HTML(html_text)
-        if not StringUtils.is_valid_html_element(html):
-            return False
-        # 站点签到支持的识别XPATH
-        xpaths = [
-            '//a[@id="signed"]',
-            '//a[contains(@href, "attendance")]',
-            '//a[contains(text(), "签到")]',
-            '//a/b[contains(text(), "签 到")]',
-            '//span[@id="sign_in"]/a',
-            '//a[contains(@href, "addbonus")]',
-            '//input[@class="dt_button"][contains(@value, "打卡")]',
-            '//a[contains(@href, "sign_in")]',
-            '//a[contains(@onclick, "do_signin")]',
-            '//a[@id="do-attendance"]',
-            '//shark-icon-button[@href="attendance.php"]'
-        ]
-        for xpath in xpaths:
-            if html.xpath(xpath):
+        try:
+            if not StringUtils.is_valid_html_element(html):
                 return False
-
-        return True
+            # 站点签到支持的识别XPATH
+            xpaths = [
+                '//a[@id="signed"]',
+                '//a[contains(@href, "attendance")]',
+                '//a[contains(text(), "签到")]',
+                '//a/b[contains(text(), "签 到")]',
+                '//span[@id="sign_in"]/a',
+                '//a[contains(@href, "addbonus")]',
+                '//input[@class="dt_button"][contains(@value, "打卡")]',
+                '//a[contains(@href, "sign_in")]',
+                '//a[contains(@onclick, "do_signin")]',
+                '//a[@id="do-attendance"]',
+                '//shark-icon-button[@href="attendance.php"]'
+            ]
+            for xpath in xpaths:
+                if html.xpath(xpath):
+                    return False
+            return True
+        finally:
+            if html is not None:
+                del html

version.py

@@ -1,2 +1,2 @@
-APP_VERSION = 'v2.6.2'
-FRONTEND_VERSION = 'v2.6.2'
+APP_VERSION = 'v2.6.3'
+FRONTEND_VERSION = 'v2.6.3'