fix v tail - bug

bilix/api/bilibili.py

@@ -150,11 +150,11 @@ async def get_up_info(client: httpx.AsyncClient, url_or_mid: str, pn=1, ps=30, o
 @dataclass
 class VideoInfo:
     title: str
-    h1_title: str
+    h1_title: str  # for bv same to title, but for tv or bangumi title will be more specific
     aid: Union[str, int]
     cid: Union[str, int]
     p: int
-    pages: Sequence[Sequence[str]]
+    pages: Sequence[Sequence[str]]  # [[p_name, p_url], ...]
     img_url: str
     status: dict
     bvid: str = None
@@ -197,8 +197,8 @@ async def get_video_info(client: httpx.AsyncClient, url) -> VideoInfo:
         base_url = url.split('?')[0]
         for idx, i in enumerate(init_info['videoData']['pages']):
             p_url = f"{base_url}?p={idx + 1}"
-            add_name = f"P{idx + 1}-{i['part']}" if len(init_info['videoData']['pages']) > 1 else ''
-            pages.append([add_name, p_url])
+            p_name = f"P{idx + 1}-{i['part']}" if len(init_info['videoData']['pages']) > 1 else ''
+            pages.append([p_name, p_url])
     elif 'initEpList' in init_info:  # 动漫，电视剧，电影
         stat = init_info['mediaInfo']['stat']
         status = {
@@ -218,8 +218,8 @@ async def get_video_info(client: httpx.AsyncClient, url) -> VideoInfo:
         title = legal_title(re.search('property="og:title" content="([^"]*)"', res.text).groups()[0])
         for idx, i in enumerate(init_info['initEpList']):
             p_url = i['link']
-            add_name = i['title']
-            pages.append([add_name, p_url])
+            p_name = i['title']
+            pages.append([p_name, p_url])
     else:
         raise AttributeError("未知类型")
 

bilix/download/downloader_bilibili.py

@@ -312,23 +312,20 @@ async def get_series(self, url: str, quality: Union[str, int] = 0, image=False,
             hierarchy = self._make_hierarchy_dir(hierarchy, title)
         else:
             hierarchy = hierarchy if type(hierarchy) is str else ''  # incase hierarchy is False
-        cors = [self.get_video(p_url, quality, add_name,
-                               image=image,
-                               subtitle=subtitle, dm=dm, only_audio=only_audio, codec=codec, hierarchy=hierarchy,
-                               video_info=video_info if idx == p else None)
-                for idx, (add_name, p_url) in enumerate(pages)]
+        cors = [self.get_video(p_url, quality=quality, image=image, subtitle=subtitle, dm=dm, only_audio=only_audio,
+                               codec=codec, hierarchy=hierarchy, video_info=video_info if idx == p else None)
+                for idx, (_, p_url) in enumerate(pages)]
         if p_range:
             cors = cors_slice(cors, p_range)
         await asyncio.gather(*cors)
 
-    async def get_video(self, url: str, quality: Union[str, int] = 0, add_name='', image=False, subtitle=False,
+    async def get_video(self, url: str, quality: Union[str, int] = 0, image=False, subtitle=False,
                         dm=False, only_audio=False, codec: str = '', hierarchy: str = '', video_info=None):
         """
         下载单个视频
 
         :param url: 视频的url
         :param quality: 画面质量，0为可以观看的最高画质，越大质量越低，超过范围时自动选择最低画质，或者直接使用字符串指定'1080p'等名称
-        :param add_name: 给文件的额外添加名，用户请直接保持默认
         :param image: 是否下载封面
         :param subtitle: 是否下载字幕
         :param dm: 是否下载弹幕
@@ -345,19 +342,20 @@ async def get_video(self, url: str, quality: Union[str, int] = 0, add_name='', i
                 except AttributeError as e:
                     logger.warning(f'{url} {e}')
                     return
-            title = video_info.h1_title
-            title = legal_title(title, add_name)
-            video_info.title = title  # update video_info title
+            p = video_info.p
+            p_name = video_info.pages[p][0]
+            # join p_name and title
+            title = legal_title(video_info.h1_title, p_name)
             img_url = video_info.img_url
             if not video_info.dash:
-                logger.warning(f'{video_info.title} 需要大会员或该地区不支持')
+                logger.warning(f'{title} 需要大会员或该地区不支持')
                 return
             # choose video quality
             try:
                 video, audio = choose_quality(video_info, quality, codec)
             except ValueError:
                 logger.warning(
-                    f"{video_info.title} 清晰度<{quality}> 编码<{codec}>不可用，请检查输入是否正确或是否需要大会员")
+                    f"{title} 清晰度<{quality}> 编码<{codec}>不可用，请检查输入是否正确或是否需要大会员")
                 return
 
             file_dir = f'{self.videos_dir}/{hierarchy}' if hierarchy else self.videos_dir
@@ -414,12 +412,10 @@ async def get_dm(self, url, update=False, convert_func=None, hierarchy: str = ''
         """
         if not video_info:
             video_info = await api.get_video_info(self.client, url)
-        title = video_info.title
         aid, cid = video_info.aid, video_info.cid
-
         file_dir = f'{self.videos_dir}/{hierarchy}' if hierarchy else self.videos_dir
         file_type = '.' + ('bin' if not convert_func else convert_func.__name__.split('2')[-1])
-        file_name = f"{title}-弹幕{file_type}"
+        file_name = legal_title(video_info.h1_title, video_info.pages[video_info.p][0], "弹幕") + file_type
         file_path = f'{file_dir}/{file_name}'
         if not update and os.path.exists(file_path):
             logger.info(f"[green]已存在[/green] {file_name}")
@@ -436,32 +432,29 @@ async def get_dm(self, url, update=False, convert_func=None, hierarchy: str = ''
         logger.info(f"[cyan]已完成[/cyan] {file_name}")
         return file_path
 
-    async def get_subtitle(self, url, convert=True, hierarchy: str = '', video_info=None):
+    async def get_subtitle(self, url, convert_func=json2srt, hierarchy: str = '', video_info=None):
         """
         获取某个视频的字幕文件
 
         :param url: 视频url
-        :param convert: 是否转换成srt
+        :param convert_func: function used to convert original subtitle text
         :param hierarchy:
         :param video_info: 额外数据，提供则不再访问前端
         :return:
         """
         if not video_info:
             video_info = await api.get_video_info(self.client, url)
         p, cid = video_info.p, video_info.cid
-        title = video_info.title
-        add_name = video_info.pages[p][0]
-        title = legal_title(title, add_name)
+        p_name = video_info.pages[p][0]
         try:
             subtitles = await api.get_subtitle_info(self.client, video_info.bvid, cid)
         except AttributeError as e:
             logger.warning(f'{url} {e}')
             return
         cors = []
         for sub_url, sub_name in subtitles:
-            file_name = f"{title}-{sub_name}"
-            cors.append(self._get_static(sub_url, file_name, convert_func=json2srt if convert else None,
-                                         hierarchy=hierarchy))
+            file_name = legal_title(video_info.h1_title, p_name, sub_name)
+            cors.append(self._get_static(sub_url, file_name, convert_func=convert_func, hierarchy=hierarchy))
         paths = await asyncio.gather(*cors)
         return paths
 

bilix/utils.py

@@ -56,16 +56,17 @@ async def merge_files(file_list: Sequence[str], new_name: str):
     os.rename(first_file, new_name)
 
 
-def legal_title(*parts: str, join_str: str = '-', max_length=100):
+def legal_title(*parts: str, join_str: str = '-', max_length=150):
     """
-    join several string parts to os illegal file/dir name (no illegal character and not too long)
+    join several string parts to os illegal file/dir name (no illegal character and not too long).
+    auto skip empty.
 
     :param parts:
     :param join_str: the string to join each part
     :param max_length: joined file name max length, tail will be truncated if overly long
     :return:
     """
-    return f'{join_str.join(map(replace_illegal, parts)):.{max_length}}'
+    return f'{join_str.join(filter(lambda x: len(x) > 0, map(replace_illegal, parts))):.{max_length}}'
 
 
 def replace_illegal(s: str):